# AI Agent Security Solutions: every page as markdown Sixteen AI agent security solutions scored on five layers and six capability criteria, read 16 Nov 2026, with a layer view for each place agents run. --- # AI agent security solutions, ranked by layer URL: https://aiagentsecuritysolutions.com/ Updated: 2026-11-16 Sixteen AI agent security solutions, scored 0 to 100 on 16 Nov 2026 from public evidence. Operant AI leads the site ranking at 64.9. Pick a layer on the map below, endpoint, browser, cloud workspace, identity or gateway, and the table re-ranks for where your agents run. ## The short answer Operant AI leads the site ranking at 64.9 of 100. Bay is 4th at 62.0. Operant AI is ahead mainly on coverage: gateway layer (7 against 0, worth 4.2 points at weight 6). ## Pick a layer AI agents run in five places. Each layer has its own ranked table, built from the same scores with the weight moved to that layer. With scripts on, the buttons show one layer at a time. EndpointBrowserCloud workspaceIdentityGateway ### Endpoint layer: ranked solutions Bay leads the view "Ranked for the endpoint layer" at 71.0 of 100. Microsoft follows at 69.0. Bay is ahead of Microsoft mainly on stops the action (9 against 6, worth 6.0 points at weight 20). Bay is a young company that left stealth in 2026, publishes no price and has little independent testing behind its claims. **Ranked for the endpoint layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 30) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Bay](/vendors/bay/) | 71.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | 2nd | [Microsoft](/vendors/microsoft/) | 69.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 3rd | [Operant AI](/vendors/operant-ai/) | 68.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 4th | [Straiker](/vendors/straiker/) | 65.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | joint 5th | [CrowdStrike](/vendors/crowdstrike/) | 64.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | joint 5th | [Pillar Security](/vendors/pillar-security/) | 64.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 7th | [Cisco](/vendors/cisco/) | 63.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | joint 8th | [Aim Security](/vendors/aim-security/) | 62.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | joint 8th | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | joint 10th | [Lasso Security](/vendors/lasso-security/) | 62.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | joint 10th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 12th | [Noma Security](/vendors/noma-security/) | 57.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 13th | [Check Point](/vendors/check-point/) | 55.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 14th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 53.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 15th | [WitnessAI](/vendors/witnessai/) | 52.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 51.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | ### Browser layer: ranked solutions Straiker leads the view "Ranked for the browser layer" at 65.5 of 100. Bay is 2nd at 65.0. Straiker is ahead mainly on coverage: browser layer (7 against 6, worth 3.0 points at weight 30). **Ranked for the browser layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 30) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Straiker](/vendors/straiker/) | 65.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 2nd | [Bay](/vendors/bay/) | 65.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | joint 3rd | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | joint 3rd | [Operant AI](/vendors/operant-ai/) | 62.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 5th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 62.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 6th | [Aim Security](/vendors/aim-security/) | 59.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 7th | [CrowdStrike](/vendors/crowdstrike/) | 58.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 8th | [Check Point](/vendors/check-point/) | 58.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 9th | [Microsoft](/vendors/microsoft/) | 51.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 10th | [Lasso Security](/vendors/lasso-security/) | 50.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | 11th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 48.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | | 12th | [Cisco](/vendors/cisco/) | 45.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 13th | [Zenity](/vendors/zenity/) | 44.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | joint 14th | [Pillar Security](/vendors/pillar-security/) | 43.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | joint 14th | [WitnessAI](/vendors/witnessai/) | 43.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [Noma Security](/vendors/noma-security/) | 42.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | ### Cloud workspace layer: ranked solutions Operant AI and Straiker are joint 1st in the view "Ranked for the cloud workspace layer" at 68.5 of 100. Bay is joint 5th at 62.0. Operant AI is ahead mainly on coverage: cloud workspace layer (8 against 5, worth 9.0 points at weight 30). **Ranked for the cloud workspace layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 30) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | joint 1st | [Operant AI](/vendors/operant-ai/) | 68.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | joint 1st | [Straiker](/vendors/straiker/) | 68.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 3rd | [Palo Alto Networks](/vendors/palo-alto-networks/) | 68.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 4th | [Microsoft](/vendors/microsoft/) | 66.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | joint 5th | [Bay](/vendors/bay/) | 62.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | joint 5th | [Lasso Security](/vendors/lasso-security/) | 62.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | joint 5th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 8th | [Check Point](/vendors/check-point/) | 61.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 9th | [Noma Security](/vendors/noma-security/) | 60.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 10th | [Cisco](/vendors/cisco/) | 57.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 11th | [Aim Security](/vendors/aim-security/) | 53.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 12th | [Island](/vendors/island/) | 50.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | 13th | [Pillar Security](/vendors/pillar-security/) | 49.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 14th | [CrowdStrike](/vendors/crowdstrike/) | 46.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 15th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 45.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | | 16th | [WitnessAI](/vendors/witnessai/) | 43.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | ### Identity layer: ranked solutions CrowdStrike leads the view "Ranked for the identity layer" at 67.5 of 100. Bay is 4th at 65.0. CrowdStrike is ahead mainly on coverage: identity layer (8 against 6, worth 6.0 points at weight 30). **Ranked for the identity layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 30) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [CrowdStrike](/vendors/crowdstrike/) | 67.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 2nd | [Microsoft](/vendors/microsoft/) | 66.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 3rd | [Straiker](/vendors/straiker/) | 65.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 4th | [Bay](/vendors/bay/) | 65.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | 5th | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | 6th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 7th | [Cisco](/vendors/cisco/) | 60.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 8th | [Operant AI](/vendors/operant-ai/) | 59.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 9th | [Noma Security](/vendors/noma-security/) | 57.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 10th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 56.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 11th | [Lasso Security](/vendors/lasso-security/) | 53.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | 12th | [Pillar Security](/vendors/pillar-security/) | 49.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 13th | [Aim Security](/vendors/aim-security/) | 47.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 14th | [Check Point](/vendors/check-point/) | 46.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 15th | [WitnessAI](/vendors/witnessai/) | 43.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 42.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | ### Gateway layer: ranked solutions Palo Alto Networks leads the view "Ranked for the gateway layer" at 68.0 of 100. Bay is 16th at 47.0. Palo Alto Networks is ahead mainly on coverage: gateway layer (8 against 0, worth 24.0 points at weight 30). **Ranked for the gateway layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 30) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Palo Alto Networks](/vendors/palo-alto-networks/) | 68.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 2nd | [Operant AI](/vendors/operant-ai/) | 65.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 3rd | [Lasso Security](/vendors/lasso-security/) | 65.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | 4th | [CrowdStrike](/vendors/crowdstrike/) | 64.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 5th | [Microsoft](/vendors/microsoft/) | 63.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 6th | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | 7th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 8th | [WitnessAI](/vendors/witnessai/) | 61.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 9th | [Check Point](/vendors/check-point/) | 61.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 10th | [Cisco](/vendors/cisco/) | 60.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 11th | [Aim Security](/vendors/aim-security/) | 59.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 12th | [Pillar Security](/vendors/pillar-security/) | 58.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 13th | [Noma Security](/vendors/noma-security/) | 57.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 14th | [Straiker](/vendors/straiker/) | 53.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 15th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 48.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | | 16th | [Bay](/vendors/bay/) | 47.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | ## The ranking [The full ranking](/ranking/) has the same table with re-rank sliders. Each score links to the evidence on the vendor's page. **Site ranking: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 6) | Coverage: browser layer (weight 6) | Coverage: cloud workspace layer (weight 6) | Coverage: identity layer (weight 6) | Coverage: gateway layer (weight 6) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Operant AI](/vendors/operant-ai/) | 64.9 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 2nd | [Straiker](/vendors/straiker/) | 63.7 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 3rd | [Microsoft](/vendors/microsoft/) | 63.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 4th | [Bay](/vendors/bay/) | 62.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | 5th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 61.4 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 6th | [CrowdStrike](/vendors/crowdstrike/) | 60.3 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 7th | [Island](/vendors/island/) | 60.1 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | joint 8th | [Lasso Security](/vendors/lasso-security/) | 58.4 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | joint 8th | [Zenity](/vendors/zenity/) | 58.4 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 10th | [Cisco](/vendors/cisco/) | 57.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 11th | [Aim Security](/vendors/aim-security/) | 56.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 12th | [Check Point](/vendors/check-point/) | 56.2 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 13th | [Noma Security](/vendors/noma-security/) | 54.6 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 14th | [Pillar Security](/vendors/pillar-security/) | 53.1 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 15th | [WitnessAI](/vendors/witnessai/) | 48.9 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 46.8 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | ## Who should pick what With no particular situation, the published ranking applies and Operant AI comes first. For a specific situation the weights change, and so can the order: - **[Ranked for the endpoint layer](/best-for/layer-endpoint/)**: pick Bay (71.0). Bay is 1st at 71.0. A buyer securing agents at the endpoint layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the browser layer](/best-for/layer-browser/)**: pick Straiker (65.5). Bay is 2nd at 65.0. A buyer securing agents at the browser layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the cloud workspace layer](/best-for/layer-cloud/)**: pick Operant AI and Straiker (68.5, joint 1st). Bay is joint 5th at 62.0. A buyer securing agents at the cloud workspace layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the identity layer](/best-for/layer-identity/)**: pick CrowdStrike (67.5). Bay is 4th at 65.0. A buyer securing agents at the identity layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the gateway layer](/best-for/layer-gateway/)**: pick Palo Alto Networks (68.0). Bay is 16th at 47.0. A buyer securing agents at the gateway layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. ## What would change this ranking **Top pair.** Straiker rising from 7 to 8 on stops the action (weight 20, so each point moves a total by 2.0) would put Straiker above Operant AI; the gap is 1.2 points. **Bay's next place.** Bay rising from 9 to 10 on stops the action (weight 20, so each point moves a total by 2.0) would put Bay above Microsoft; the gap is 1.0 points. A score changes only when a sourced fact changes, and the weights were fixed before the scores. The [method page](/method/) lists the weights, anchors and rules. ## What are the best AI agent security solutions? On the site ranking, Operant AI is first with 64.9 of 100, and Bay is 4th with 62.0. The ranking sums five layer scores (6 points each) and six capability scores: stopping the action, visibility, deployment, policy, transparency and proof. The order changes with the layer you secure first. Each layer view gives its own layer 30 points instead of 6, so a team that starts at the browser or the gateway sees a different leader. The layer map on this page switches between the five views. ## Which layer of AI agent security do you need? You need the layer where your agents run today, and most teams run agents in more than one place. Employees run desktop and coding agents on laptops, browser agents inside the browser, SaaS agents in cloud workspaces, and every agent carries credentials and calls tools or models through some path. The five layers are the endpoint, the browser, the cloud workspace, identity and the gateway. Choose the layer map entry that matches your first deployment and read the vendors in that view. The guide on layers explains what each layer can and cannot see. - Endpoint: desktop agents, coding agents, command-line agents and local MCP servers. - Browser: browser agents, AI browsers and AI extensions. - Cloud workspace: agents built or hosted in SaaS and cloud platforms such as Copilot Studio, Agentforce and Bedrock. - Identity: agent identities, credentials, tokens and permissions. - Gateway: a proxy or API control point between agents and models, tools or MCP servers. ## Which AI agent security platform covers the most layers? Operant AI is the only solution in this set scored at a working level (5 or more) on all five layers; eight others reach four, including Bay, Straiker, Island, Zenity, Microsoft, CrowdStrike, Cisco and Noma Security. Coverage is read from vendor documentation, so each claim is tied to a source on the vendor page. More layers do not mean a better fit. A solution that blocks agent actions on the endpoint may have no gateway, and a gateway-first product may see nothing on a laptop. The Layer Coverage Map lays out every vendor by layer, and the Agent Control Point Map shows where in an agent's run each one acts. ## How was this ranking built? Every vendor got the same research plan on 16 Nov 2026: vendor pages, documentation, pricing and release notes, plus at least two third-party hosts per vendor. The result is 1,469 quote-checked facts across 32 vendor files; the 16 files for this site hold 902 of them. A scorer worked from the evidence with each vendor's own name, parent and hosts removed from its own record and replaced by a code, and a blind reviewer checked the scores. Weights were committed before any evidence or score existed and have not changed. The method page lists every criterion, the anchors and the rules. ## Latest news, guides and data - [CrowdStrike Falcon AI Gateway reaches general availability](/news/crowdstrike-falcon-ai-gateway-generally-available/) (2026-11-13) - [CrowdStrike adds Seraphic browser security to Falcon](/news/crowdstrike-seraphic-enterprise-browser-falcon/) (2026-11-04) - [Cato Networks launches Cato AI Endpoint for local agents](/news/cato-ai-endpoint-launch/) (2026-10-27) - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [AI Agent Security for Multi-Agent Systems](/guides/ai-agent-security-for-multi-agent-systems/) - [AI Agent Security for Small Teams: Which Layer First](/guides/ai-agent-security-for-small-teams-which-layer-first/) - [Layer Coverage Map](/layer-coverage-map/), with [CSV and JSON downloads](/datasets/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/), with [CSV and JSON downloads](/datasets/agent-control-point-map/) ## Related pages - [Full ranking with re-rank sliders](/ranking/) - [How the scores and weights work](/method/) - [Vendor profiles with evidence](/vendors/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) - [Guides](/guides/) - [Frequently asked questions](/faq/) --- # AI agent security solutions ranking URL: https://aiagentsecuritysolutions.com/ranking/ Updated: 2026-11-16 Operant AI ranks first among sixteen AI agent security solutions with 64.9 of 100 on 16 Nov 2026. Bay is 4th with 62.0. Totals come from five layer scores at 6 points each plus stopping the action (20), visibility (15), deployment (10), policy (10), transparency (10) and proof (5). ## The short answer Operant AI leads the site ranking at 64.9 of 100. Bay is 4th at 62.0. Operant AI is ahead mainly on coverage: gateway layer (7 against 0, worth 4.2 points at weight 6). **Site ranking: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 6) | Coverage: browser layer (weight 6) | Coverage: cloud workspace layer (weight 6) | Coverage: identity layer (weight 6) | Coverage: gateway layer (weight 6) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Operant AI](/vendors/operant-ai/) | 64.9 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 2nd | [Straiker](/vendors/straiker/) | 63.7 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 3rd | [Microsoft](/vendors/microsoft/) | 63.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 4th | [Bay](/vendors/bay/) | 62.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | 5th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 61.4 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 6th | [CrowdStrike](/vendors/crowdstrike/) | 60.3 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 7th | [Island](/vendors/island/) | 60.1 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | joint 8th | [Lasso Security](/vendors/lasso-security/) | 58.4 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | joint 8th | [Zenity](/vendors/zenity/) | 58.4 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 10th | [Cisco](/vendors/cisco/) | 57.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 11th | [Aim Security](/vendors/aim-security/) | 56.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 12th | [Check Point](/vendors/check-point/) | 56.2 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 13th | [Noma Security](/vendors/noma-security/) | 54.6 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 14th | [Pillar Security](/vendors/pillar-security/) | 53.1 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 15th | [WitnessAI](/vendors/witnessai/) | 48.9 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 46.8 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | ## Who should pick what With no particular situation, the published ranking applies and Operant AI comes first. For a specific situation the weights change, and so can the order: - **[Ranked for the endpoint layer](/best-for/layer-endpoint/)**: pick Bay (71.0). Bay is 1st at 71.0. A buyer securing agents at the endpoint layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the browser layer](/best-for/layer-browser/)**: pick Straiker (65.5). Bay is 2nd at 65.0. A buyer securing agents at the browser layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the cloud workspace layer](/best-for/layer-cloud/)**: pick Operant AI and Straiker (68.5, joint 1st). Bay is joint 5th at 62.0. A buyer securing agents at the cloud workspace layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the identity layer](/best-for/layer-identity/)**: pick CrowdStrike (67.5). Bay is 4th at 65.0. A buyer securing agents at the identity layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the gateway layer](/best-for/layer-gateway/)**: pick Palo Alto Networks (68.0). Bay is 16th at 47.0. A buyer securing agents at the gateway layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. ## What would change this ranking **Top pair.** Straiker rising from 7 to 8 on stops the action (weight 20, so each point moves a total by 2.0) would put Straiker above Operant AI; the gap is 1.2 points. **Bay's next place.** Bay rising from 9 to 10 on stops the action (weight 20, so each point moves a total by 2.0) would put Bay above Microsoft; the gap is 1.0 points. A score changes only when a sourced fact changes, and the weights were fixed before the scores. The [method page](/method/) lists the weights, anchors and rules. ## Which AI agent security solution ranks first? Operant AI ranks first on the site ranking with 64.9 of 100. The ranking sums weighted scores from 0 to 10 on eleven criteria, divided by 10, and equal totals print as a joint rank. Gaps between neighbors are often small, so read the layer views before you shortlist. A vendor in the middle of the table can lead the one layer you need. ## How do AI agent security tools differ by layer? They differ in where they can see and where they can act. Endpoint tools sit on the device and hook tool calls, browser tools sit in the extension or browser, cloud workspace tools connect by API to SaaS and cloud agent platforms, identity tools manage agent credentials, and gateways sit on the traffic path. No single position sees every agent. The layer views re-weight the ranking so the chosen layer carries 30 points. Bay leads the endpoint view, Straiker the browser view, Operant AI and Straiker share the cloud view, CrowdStrike leads identity and Palo Alto Networks leads the gateway view. ## What does the ranking leave out? It scores public evidence only: product pages, documentation, pricing pages, release notes and third-party write-ups. A feature with no public description scores 0 on that point and is written as none found on 16 Nov 2026. It does not test products in a lab. A vendor claim alone is capped at 7 of 10 until a second source confirms it, so a score of 8 or more means a third-party or directly observed source backs the claim. ## Related pages - [Method: weights, anchors and rules](/method/) - [Rankings by situation](/best-for/) - [Vendor profiles](/vendors/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) - [Guides for buyers](/guides/) - [Glossary](/glossary/) --- # How AI agent security solutions are scored URL: https://aiagentsecuritysolutions.com/method/ Updated: 2026-11-16 One research plan covered every vendor on 16 Nov 2026; the sixteen evidence files here hold 902 facts, and each quote was checked against its page. One scorer scored blind, a second blind reviewer checked, and weights were committed before any evidence. The five layers weigh 6 each. ## The weights The weights were fixed on 2026-11-16, before any score existed, and have not changed since. **Criteria and weights (out of 100)** | Criterion | What it measures | Weight | |---|---|---| | Coverage: endpoint layer | Agents running on laptops and developer machines (desktop agents, coding agents, command-line agents, local MCP servers), as documented | 6 | | Coverage: browser layer | Browser agents, AI browsers and AI extensions, as documented | 6 | | Coverage: cloud workspace layer | Agents built or hosted in SaaS and cloud platforms (Microsoft Copilot Studio, Salesforce Agentforce, ServiceNow, AWS Bedrock agents and similar), as documented | 6 | | Coverage: identity layer | Agent identities, credentials, tokens and permissions: inventory, least privilege, revocation, as documented | 6 | | Coverage: gateway layer | A gateway, proxy or API control point between agents and models, tools or MCP servers, as documented | 6 | | Stops the action | Blocking a harmful agent action before it completes, with the action types named, against detection or alerting after the fact | 20 | | Visibility | An inventory of agents and what each can reach (tools, data, permissions) with activity logs | 15 | | Deployment | How it deploys (agentless, API, extension, agent, proxy), what it needs, time to deploy as published | 10 | | Policy and governance | The policy model, audit trail, and integrations with SIEM, EDR, the identity provider and ticketing | 10 | | Transparency | Public documentation readable without sign-in, a published price, published research | 10 | | Proof | Named customers, funding, independent analyst coverage | 5 | ## Weights by situation **Weights used for each situation view** | Situation | Coverage: endpoint layer | Coverage: browser layer | Coverage: cloud workspace layer | Coverage: identity layer | Coverage: gateway layer | Stops the action | Visibility | Deployment | Policy and governance | Transparency | Proof | |---|---|---|---|---|---|---|---|---|---|---|---| | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 30 | 0 | 0 | 0 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | | [Ranked for the browser layer](/best-for/layer-browser/) | 0 | 30 | 0 | 0 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 0 | 0 | 30 | 0 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 0 | 0 | 0 | 30 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 0 | 0 | 0 | 0 | 30 | 20 | 15 | 10 | 10 | 10 | 5 | ## Score anchors - **0**: none found on the read date - **1-3**: mentioned, partial, or limited to one surface, one layer or one action type - **4-6**: documented and working, with gaps the documentation shows - **7**: the most a vendor claim or vendor documentation alone can earn: detailed, specific, covering the whole criterion - **8-10**: detailed and confirmed by a second source (third-party write-up, or observed directly in public documentation, code, a marketplace listing or a release record) ## Rules - Every vendor is scored 0 to 10 on every criterion, whole numbers, from the evidence file only, by a scorer that works from coded packets with every vendor name removed. - A vendor claim scores at most 7 until a second source confirms it (kind third-party or observed). - None found scores 0 and is written "none found on the read date". - Breadth facts (module counts, integration counts, headcount, funding) inform Proof only, which weighs 5, so they carry at most 10 percent of any ranking. - The total is the weighted sum of the scores divided by 10 (0 to 100 scale, published to one decimal). - Equal totals print as a joint rank (for example joint 2nd); inside a joint rank vendors are listed alphabetically for display only. - Once the first score exists, no weight and no criterion changes. A score changes only when a sourced fact changes, logged with the fact. ## How are AI agent security solutions scored? Each vendor is scored 0 to 10, in whole numbers, on eleven criteria, from the evidence file only. The total is the weighted sum divided by 10, giving a 0 to 100 scale printed to one decimal. The same research plan ran for every vendor on 16 Nov 2026: vendor pages, documentation, pricing pages and release notes, plus at least two third-party hosts per vendor. It produced 1,469 facts across 32 vendor files. Every fact carries a source URL, a short verbatim quote and a read date, and each quote was checked against its page. ## Why do the five layers weigh 6 each? The five layer criteria (endpoint, browser, cloud workspace, identity and gateway) weigh 6 each, 30 in all, so no single place agents run can dominate the site ranking. That equal split is a choice made before scoring, and it is recorded in the project decisions log. Each layer view moves all 30 points onto its own layer and sets the other four layers to 0. The other six criteria keep their weights: stopping the action 20, visibility 15, deployment 10, policy 10, transparency 10 and proof 5. ## How does blind scoring work? One scorer reads the evidence file for each vendor with each vendor's own name, parent and hosts removed from its own record and replaced by a code, and scores every criterion with a short written reason and the fact ids behind it. A second reviewer, also blind, checks the scores against the same facts and flags any that the evidence does not support. Scores change only when a sourced fact changes, and the change is logged with the fact. ## What do the score anchors mean? A 0 means none found on the read date, 1 to 3 means mentioned or limited to one surface, one layer or one action type, and 4 to 6 means documented and working with gaps the documentation shows. A 7 is the most a vendor claim or vendor documentation alone can earn. A score of 8 to 10 needs detail confirmed by a second source: a third-party write-up, or direct observation of public documentation, code, a marketplace listing or a release record. ## How are corrections handled? A vendor or reader can send a sourced correction to the address on the corrections page. If the source changes a fact, the fact, any score that rests on it and the dated changelog entry are updated together. Weights and criteria stay fixed once the first score exists, so a correction can move a score but not the rules. ## Changelog - 2026-11-16: weights and criteria fixed. - 2026-11-16: evidence read for every vendor by one research plan. - 2026-11-16: scores published (blind: one scorer, then one blind reviewer). - 2026-11-16: this site built from the committed scores. ## Related pages - [The ranking these weights produce](/ranking/) - [Vendor profiles with evidence per score](/vendors/) - [Editorial policy](/editorial-policy/) - [Corrections](/corrections/) - [Glossary](/glossary/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # AI agent security vendors URL: https://aiagentsecuritysolutions.com/vendors/ Updated: 2026-11-16 Sixteen AI agent security vendors are profiled here, from startups such as Bay, Operant AI and Straiker to platform companies such as Microsoft, CrowdStrike, Cisco and Palo Alto Networks. Each profile lists scores, the reason for each score and the sourced facts behind it, read on 16 Nov 2026. ## All 16 vendors **Vendors in the ranking, with their rank and total** | Vendor | Rank | Total (of 100) | Product assessed | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | 11th | 56.5 | Cato AI Security for Agents with Cato AI Endpoint | | [Bay](/vendors/bay/) | 4th | 62.0 | Bay platform: Endpoint, Cloud Workspaces and Browser, with AI Security Posture and Agent Runtime Enforcement | | [Check Point](/vendors/check-point/) | 12th | 56.2 | Check Point AI Agent Security (discovery, risk assessment, runtime via AI Guardrails) plus Workforce AI Security for endpoint and browser coverage | | [Cisco](/vendors/cisco/) | 10th | 57.0 | Cisco AI Defense (runtime guardrails, AI Cloud Visibility, Explorer red teaming) with open-source DefenseClaw and Duo IAM agent registration (in development) | | [CrowdStrike](/vendors/crowdstrike/) | 6th | 60.3 | Falcon Guardian (AIDR) with Falcon AI Gateway and AI Kill Switch, plus Agentic Identity Provider and Falcon Seraphic Enterprise Browser for identity and browser layers | | [Island](/vendors/island/) | 7th | 60.1 | Island Enterprise Agentic Control Plane: Agentic Endpoint Posture, Agentic Identity, AI Runtime Protection, Island MCP Gateway, Enterprise Browser for Agents, Island Extension and Island Desktop | | [Lasso Security](/vendors/lasso-security/) | joint 8th | 58.4 | Lasso AI Security Platform: AI Agents Security, Discovery & AI-BOM, AI Usage Control, AI Gateway Security | | [Microsoft](/vendors/microsoft/) | 3rd | 63.0 | Microsoft Agent 365 with Defender for Endpoint and Defender AI agent protection, Microsoft Entra Agent ID and Conditional Access for agents, Purview | | [Noma Security](/vendors/noma-security/) | 13th | 54.6 | Noma Agent Security and Governance Platform: AI-SPM, Agent Access Control, AI-DR for endpoint, SaaS and cloud agents | | [Operant AI](/vendors/operant-ai/) | 1st | 64.9 | Agent Protector, Endpoint Protector, MCP Gateway, AI Gatekeeper and Browser Protection | | [Palo Alto Networks](/vendors/palo-alto-networks/) | 5th | 61.4 | Prisma AIRS (AI Gateway, SaaS Agent Security, Agent Identity Security, AI Runtime Security) with Prisma Browser | | [Pillar Security](/vendors/pillar-security/) | 14th | 53.1 | Agentic Endpoint, Runtime Guardrails and AI Discovery & Posture | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 16th | 46.8 | Prompt Security Agentic AI Security, with the MCP Gateway and Amazon Bedrock AgentCore integration (SentinelOne) | | [Straiker](/vendors/straiker/) | 2nd | 63.7 | Straiker platform: Discover AI (inventory and posture), Defend AI (runtime guardrails) and the Agentic Kill Switch; Ascend AI red teaming supports them | | [WitnessAI](/vendors/witnessai/) | 15th | 48.9 | WitnessAI Agentic Control (Observe, Control and Protect for agents, MCP Catalog, runtime protection REST APIs) | | [Zenity](/vendors/zenity/) | joint 8th | 58.4 | Zenity AI Agent Security and Governance Platform: AI Observability, AISPM, AI Exposure Management, Boundaries (runtime enforcement), Agentic Identity, MCP Security and AIDR | ## Related pages - [Ranking of all vendors](/ranking/) - [How vendors are scored](/method/) - [Best vendor by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Aim Security AI agent security URL: https://aiagentsecuritysolutions.com/vendors/aim-security/ Updated: 2026-11-16 Aim Security is now part of Cato Networks, so its agent controls appear in Cato AI Endpoint and Cato AI Security for Agents. Cato says the endpoint client finds local agents and their MCP servers and skills, blocks an unapproved tool call before it runs while the rest of the agent's work continues, and names Claude Code, Cursor and Codex. In the browser, the same rules can monitor, anonymize or block a prompt. Applications are protected through a proxy, an AI gateway integration or APIs. Cato reports more than 4,800 customers. Agent identity and token management, Copilot Studio, Agentforce and ServiceNow coverage, and a published price: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** It covers endpoint, browser and gateway layers and blocks risky agent tool calls before they run, with enforcement named for Claude Code, Cursor and Codex (aim-security-40, -41, -51). **Weakness:** None found on 2026-11-16 for agent identity and token management, named SaaS agent platforms such as Copilot Studio or Agentforce, SIEM or ticketing integrations, or a published price (aim-security-46, -48, -57, -30). ## Scores and evidence Aim Security is 11th in the overall ranking at 56.5 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Aim Security: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 7 of 10 | 6 | The vendor documents an endpoint client that finds local agents and blocks unapproved tool calls, naming Claude Code, Cursor and Codex; vendor claims only. | [aim-security-11](#fact-aim-security-11), [aim-security-39](#fact-aim-security-39), [aim-security-40](#fact-aim-security-40), [aim-security-41](#fact-aim-security-41) | | Coverage: browser layer | 6 of 10 | 6 | The vendor covers browser AI apps with a client, extensions and inline network inspection that can block prompts; browser agent detail is limited. | [aim-security-16](#fact-aim-security-16), [aim-security-42](#fact-aim-security-42), [aim-security-43](#fact-aim-security-43) | | Coverage: cloud workspace layer | 4 of 10 | 6 | The vendor governs Bedrock activity by API and discovers managed agents; Copilot Studio, Agentforce and ServiceNow coverage none found. | [aim-security-44](#fact-aim-security-44), [aim-security-45](#fact-aim-security-45), [aim-security-46](#fact-aim-security-46) | | Coverage: identity layer | 2 of 10 | 6 | The vendor mentions spotting over-privileged access; agent identity inventory, credential management and revocation none found. | [aim-security-47](#fact-aim-security-47), [aim-security-48](#fact-aim-security-48) | | Coverage: gateway layer | 6 of 10 | 6 | The vendor protects applications via proxy, AI gateway integrations or APIs and on-prem; MCP gateway specifics are limited. | [aim-security-49](#fact-aim-security-49), [aim-security-50](#fact-aim-security-50) | | Stops the action | 7 of 10 | 20 | The vendor says it blocks risky tool calls and actions before execution; the third-party item is a vendor press release, so it is not counted as standalone. | [aim-security-41](#fact-aim-security-41), [aim-security-51](#fact-aim-security-51), [aim-security-52](#fact-aim-security-52), [aim-security-53](#fact-aim-security-53), [aim-security-73](#fact-aim-security-73) | | Visibility | 7 of 10 | 15 | The vendor shows agents by user and endpoint with MCP servers, skills, data access and activity in one dashboard; vendor claims only. | [aim-security-11](#fact-aim-security-11), [aim-security-27](#fact-aim-security-27), [aim-security-54](#fact-aim-security-54), [aim-security-68](#fact-aim-security-68) | | Deployment | 5 of 10 | 10 | The client deploys through device management and goes live in weeks; supported operating systems and a time to first inventory none found. | [aim-security-19](#fact-aim-security-19), [aim-security-20](#fact-aim-security-20), [aim-security-21](#fact-aim-security-21), [aim-security-58](#fact-aim-security-58) | | Policy and governance | 5 of 10 | 10 | The vendor describes one control plane, audit trails and standards mapping; named SIEM, EDR, IdP and ticketing integrations none found. | [aim-security-28](#fact-aim-security-28), [aim-security-55](#fact-aim-security-55), [aim-security-56](#fact-aim-security-56), [aim-security-69](#fact-aim-security-69), [aim-security-57](#fact-aim-security-57) | | Transparency | 3 of 10 | 10 | Threat research is published; public documentation and a published price none found. | [aim-security-06](#fact-aim-security-06), [aim-security-30](#fact-aim-security-30), [aim-security-31](#fact-aim-security-31), [aim-security-32](#fact-aim-security-32) | | Proof | 8 of 10 | 5 | Third-party sources report large funding and the acquisition, the vendor reports over 4,800 customers and names Dayforce and MoonPay; analyst coverage is a quote only. | [aim-security-07](#fact-aim-security-07), [aim-security-08](#fact-aim-security-08), [aim-security-09](#fact-aim-security-09), [aim-security-34](#fact-aim-security-34), [aim-security-37](#fact-aim-security-37) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 11th | 56.5 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | joint 8th | 62.5 | | [Ranked for the browser layer](/best-for/layer-browser/) | 6th | 59.5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 11th | 53.5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 13th | 47.5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 11th | 59.5 | ## Product, deployment, price and owner - Product assessed: Cato AI Security for Agents with Cato AI Endpoint - Deployment: Lightweight endpoint client pushed through existing device management, plus inline network inspection of AI traffic; proxy, AI gateway or API integration for applications; on-prem or own-cloud Outposts option - Time to deploy: Cato says AI protection goes live in weeks; no operating system list found on 2026-11-16 - Price: none found on 2026-11-16; the product page offers a demo and the /pricing/ page returned 404 - Owner: Cato Networks - Website: [aim.security](https://www.aim.security) ## Facts and sources - **aim-security-01** (vendor): Cato Networks says in its own blog dated 2025-09-03 that it acquired Aim Security, and that Aim would stay available as a standalone product before converging into the Cato platform by early 2026.Quote: "In the near term, Aim will remain available as a standalone product. By early 2026, it will converge into the Cato platform"Source: [catonetworks.com](https://www.catonetworks.com/blog/securing-ai-transformation-why-cato-acquired-aim-security/), read 2026-11-16. - **aim-security-02** (third-party): SecurityWeek reported on 2025-09-03 that Cato Networks acquired Aim Security, which was founded in 2022 and came out of stealth in January 2024.Quote: "Founded in 2022 to help organizations with the secure deployment of generative-AI utilities, Aim emerged from stealth in January 2024."Source: [securityweek.com](https://www.securityweek.com/cato-networks-acquires-ai-security-firm-aim-security/), read 2026-11-16. - **aim-security-03** (third-party): In a PRNewswire release dated 2026-07-27, Cato says its Cato AI Security product combines capabilities from its Aim Security acquisition and was introduced to market in March 2026.Quote: "Cato AI Security, which converges advanced AI governance and protection capabilities from Cato's Aim Security acquisition into a fully integrated solution introduced to market in March 2026"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/cato-networks-exceeds-415m-in-arr-fueled-by-42-yoy-growth-and-enterprise-demand-302833667.html), read 2026-11-16. - **aim-security-04** (third-party): Wikipedia states that Cato Networks announced its first M&A deal, the acquisition of Aim Security, in September 2025.Quote: "Cato Networks announced its first-ever M&A deal with the acquisition of Aim Security, an AI security startup based in Israel"Source: [en.wikipedia.org](https://en.wikipedia.org/wiki/Cato_Networks), read 2026-11-16. - **aim-security-05** (vendor): Cato posts a Bloomberg summary dated 2025-09-07 saying Cato agreed to acquire Aim Security, Cato's first acquisition.Quote: "Israeli cybersecurity company Cato Networks has agreed to acquire AI startup Aim Security, marking the firm’s first acquisition."Source: [catonetworks.com](https://www.catonetworks.com/news/bloomberg-cato-networks-to-buy-aim-security-in-bid-to-make-ai-more-secure/), read 2026-11-16. - **aim-security-06** (none-found): None found on 2026-11-16: https://www.aim.security would not load (HTTP 403 Cloudflare block, also blocked via the r.jina.ai reader and /sitemap.xml), so no Aim-owned page could be read; ownership and product facts come from Cato Networks pages and third-party reports.Source: [aim.security](https://www.aim.security), read 2026-11-16. - **aim-security-07** (third-party): Wikipedia reports Cato Networks raised $50 million more in a Series G, bringing the total to $409 million at an estimated valuation above $4.8 billion.Quote: "it had raised another $50 million in its Series G round, bringing the total to $409 million at an estimated valuation of over $4.8 billion"Source: [en.wikipedia.org](https://en.wikipedia.org/wiki/Cato_Networks), read 2026-11-16. - **aim-security-08** (third-party): SecurityWeek reports Aim raised $10 million in seed funding and $18 million more in June 2025 before the acquisition, and that its researchers found the EchoLeak Microsoft 365 Copilot vulnerability (CVE-2025-32711).Quote: "Aim’s research team recently uncovered the first reported CVE of a zero-click AI vulnerability in Microsoft 365 Copilot"Source: [securityweek.com](https://www.securityweek.com/cato-networks-acquires-ai-security-firm-aim-security/), read 2026-11-16. - **aim-security-09** (third-party): Cato says on 2026-07-27 that it passed $415 million in ARR with 42% year-over-year growth and serves more than 4,800 customers.Quote: "surpassed $415 million in ARR, with 42% year-over-year (YoY) growth"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/cato-networks-exceeds-415m-in-arr-fueled-by-42-yoy-growth-and-enterprise-demand-302833667.html), read 2026-11-16. - **aim-security-11** (vendor): Cato says AI Endpoint finds local agents and their MCP servers, tools and skills on the endpoint before any traffic reaches Cato.Quote: "Cato finds local agents and their MCP servers, tools and skills on the endpoint, before any traffic reaches Cato, and flags risky ones before they act."Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-13** (vendor): Cato lists visibility use cases: discover shadow AI and coding agents on endpoints, inventory LLM and RAG apps, and map agents, MCP servers and skills.Quote: "Discover shadow AI and coding agents across endpoints"Source: [catonetworks.com](https://www.catonetworks.com/solutions/ai-security/), read 2026-11-16. - **aim-security-16** (vendor): Cato says AI Endpoint uses a lightweight client for browser AI and desktop AI apps, and also inspects browser and app AI traffic inline on the network with the same engine.Quote: "Browser AI and AI apps are also inspected inline on the network with the same AI engine, so endpoint and network AI events share one dashboard."Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-19** (vendor): Cato says AI Endpoint is pushed out with existing endpoint management tools, runs beside the EDR, and AI protection goes live in weeks.Quote: "Existing SSE, VPN and EDR keep doing their jobs, so AI protection goes live in weeks."Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-20** (vendor): Cato says the AI Endpoint client is deployed through device management like any other app, and finding agents needs no network change.Quote: "The Cato AI Endpoint client goes out through device management like any other app."Source: [catonetworks.com](https://www.catonetworks.com/blog/cato-ai-endpoint-secures-browser-ai-desktop-ai-and-coding-agents-on-every-endpoint/), read 2026-11-16. - **aim-security-21** (none-found): None found on 2026-11-16: the Cato AI Endpoint page and launch blog do not list supported operating systems or browsers, or a time to first inventory in hours or days.Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-27** (vendor): Cato says AI Endpoint shows each agent by user and endpoint with its MCP servers and skills, listing agents with risky settings first.Quote: "Agents with risky settings, such as unrestricted access to the command line, show up first, so the team knows where to start."Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-28** (vendor): Cato says AI Endpoint gives each skill a risk finding with a block, monitor or allow recommendation, and verified remote MCP servers get a badge.Quote: "The Cato AI Endpoint client gives each skill a risk finding with a block, monitor or allow recommendation. Verified remote MCP servers get a badge."Source: [catonetworks.com](https://www.catonetworks.com/blog/cato-ai-endpoint-secures-browser-ai-desktop-ai-and-coding-agents-on-every-endpoint/), read 2026-11-16. - **aim-security-29** (vendor): Cato says it maps AI usage to standards such as EU AI Act, GDPR, NIST and OWASP for auditable evidence.Quote: "Map AI usage to corporate guidelines and standards like EU AI Act, GDPR, NIST, and OWASP."Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-end-users/), read 2026-11-16. - **aim-security-30** (none-found): None found on 2026-11-16: no published price for Cato AI Security or Cato AI Endpoint on the product page, which offers "Get a product demo"; https://www.catonetworks.com/pricing/ returned HTTP 404.Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-31** (none-found): None found on 2026-11-16: the Cato support and documentation site at support.catonetworks.com returned HTTP 403 to an automated request, so public product documentation could not be read.Source: [support.catonetworks.com](https://support.catonetworks.com/), read 2026-11-16. - **aim-security-32** (vendor): Cato publishes threat research on its public blog; a 2025-04-15 Cato CTRL post covers risks of the Model Context Protocol and security practices.Quote: "As MCP-enabled tools interact with sensitive data, they increase the risk of unauthorized access."Source: [catonetworks.com](https://www.catonetworks.com/blog/cato-ctrl-exploiting-model-context-protocol-mcp/), read 2026-11-16. - **aim-security-33** (vendor): Cato offers public product pages, a white paper and a webinar on agentic AI security.Quote: "Securing agentic AI white paper"Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-34** (vendor): Cato publishes a quote from Nithin Reddy, VP Security at Dayforce, who says Dayforce deployed Cato AI extensions and put Cato AI security inline.Quote: "We prioritized the AI our employees use, so we deployed Cato AI extensions for visibility and to prevent prompt injection and data leakage."Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-end-users/), read 2026-11-16. - **aim-security-35** (vendor): Cato publishes a quote from Doug Innocenti, CIO/CSO at MoonPay, on using Cato AI Security.Quote: "Having the right AI security solution gave me the confidence to say yes to the business."Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-applications/), read 2026-11-16. - **aim-security-36** (vendor): Cato's 2026-08-03 launch release for Agentic Threat Prevention quotes Frank Dickson, group vice president at IDC, on AI-assisted attacks.Quote: "An entire attack can require less time than is required to investigate an alert"Source: [catonetworks.com](https://www.catonetworks.com/news/cato-launches-agentic-threat-prevention-to-predict-mitigate-ai-assisted-attacks/), read 2026-11-16. - **aim-security-37** (vendor): On 2026-10-06 Cato said it earned the AWS Security Competency in AI Security (Agentic AI category), one of fewer than 25 AWS partners to do so.Quote: "Cato is one of fewer than 25 AWS partners to earn the AI Security Competency"Source: [catonetworks.com](https://www.catonetworks.com/news/cato-achieves-aws-security-competency-in-ai-security/), read 2026-11-16. - **aim-security-38** (vendor): On 2026-10-27 Cato launched Cato AI Endpoint; the release quotes Alfred Sobrinho, CIO of PlayPower, and Chris DePuy of 650 Group.Quote: "“AI is creating real productivity gains across our business, but it also introduces new concerns about visibility, data protection, and governance,”"Source: [catonetworks.com](https://www.catonetworks.com/news/cato-launches-cato-ai-endpoint-secure-ai-use-beyond-browser/), read 2026-11-16. - **aim-security-39** (vendor): On 2026-10-27 Cato said Cato AI Endpoint is a client that discovers AI use, protects data and enforces policy across browsers, desktop applications and local agents.Quote: "a new endpoint client that helps organizations discover AI use, protect sensitive data, and enforce AI policies across browsers, desktop applications, and local agents"Source: [catonetworks.com](https://www.catonetworks.com/news/cato-launches-cato-ai-endpoint-secure-ai-use-beyond-browser/), read 2026-11-16. - **aim-security-40** (vendor): Cato says endpoint enforcement covers coding agents and browser AI apps, naming Claude Code, Cursor and Codex.Quote: "Covers coding agents on the endpoint and AI apps in the browser, with enforcement for Claude Code, Cursor, and Codex."Source: [catonetworks.com](https://www.catonetworks.com/solutions/ai-security/), read 2026-11-16. - **aim-security-41** (vendor): Cato says AI Endpoint blocks an unapproved tool call from a local agent on the endpoint before it runs, while the rest of the agent's work continues.Quote: "Cato AI Endpoint blocks that tool call on the endpoint before it runs, and the rest of the agent’s work carries on."Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-42** (vendor): Cato says in the browser the same rules can monitor, anonymize or block a prompt.Quote: "In the browser, the same rules can monitor a prompt, anonymize the sensitive data in the prompt, or block the prompt."Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-43** (vendor): Cato says its extensions give visibility into employee AI use and guard against prompt injection and leakage (customer Dayforce).Quote: "we deployed Cato AI extensions for visibility and to prevent prompt injection and data leakage."Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-end-users/), read 2026-11-16. - **aim-security-44** (vendor): Cato says it ingests and governs activity from Amazon Bedrock workloads through API integration without extra software in the Bedrock environment.Quote: "Through API integration, it ingests and governs activity from Bedrock-based workloads, extending visibility and policy-driven governance without requiring additional software in the Bedrock environment."Source: [catonetworks.com](https://www.catonetworks.com/news/cato-achieves-aws-security-competency-in-ai-security/), read 2026-11-16. - **aim-security-45** (vendor): Cato says it secures local, custom and managed AI agents and connects to infrastructure to discover agents across systems, tools and MCP interactions.Quote: "Cato connects to your infrastructure to discover AI agents across systems, tools, and MCP interactions."Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-46** (none-found): None found on 2026-11-16: the Cato AI Security for Agents page does not name Microsoft Copilot Studio, Salesforce Agentforce or ServiceNow agents as covered platforms.Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-47** (vendor): Cato says agents can run with excessive permissions across enterprise systems, and lists spotting over-privileged access as part of posture assessment.Quote: "Spot over-privileged access and unsafe integrations"Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-48** (none-found): None found on 2026-11-16: the Cato AI Security for Agents page does not describe agent identity inventory, credential or token management, or permission revocation.Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-49** (vendor): Cato says AI Security for Applications deploys via proxy interception, AI-gateway integrations or APIs into model platforms and developer tools.Quote: "Deploy via proxy interception, AI-gateway integrations, or APIs into model platforms and developer tools."Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-applications/), read 2026-11-16. - **aim-security-50** (vendor): Cato says applications are protected through the Cato cloud, the customer's existing AI gateway, or on-prem with Outposts.Quote: "Protects apps through the Cato cloud, the AI gateway you already run, or on-prem with Outposts."Source: [catonetworks.com](https://www.catonetworks.com/solutions/ai-security/), read 2026-11-16. - **aim-security-51** (vendor): Cato says it applies runtime controls to stop unauthorized or unsafe agent actions before execution.Quote: "Cato applies runtime controls to govern access and stop unauthorized or unsafe actions before execution."Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-52** (vendor): Cato says it inspects prompts, responses, tool calls and tool actions and blocks risky actions before they execute.Quote: "Block risky actions before they execute"Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-53** (third-party): In a PRNewswire release dated 2026-07-27 Cato says its engine inspects and enforces policy on tool inputs and outputs on all major agentic platforms.Quote: "The Cato AI engine now inspects and enforces policies on tool inputs and outputs on all major agentic platforms"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/cato-networks-exceeds-415m-in-arr-fueled-by-42-yoy-growth-and-enterprise-demand-302833667.html), read 2026-11-16. - **aim-security-54** (vendor): Cato says it discovers local, custom and managed agents with their tools, MCP servers, data access and execution paths.Quote: "Cato discovers local, custom, and managed agents across the enterprise, including their tools, MCP servers, data access, and execution paths."Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-55** (vendor): Cato says one control plane enforces consistent policy across every app and agent with a single source of truth for compliance and incident response.Quote: "One policy across the entire AI surface"Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-applications/), read 2026-11-16. - **aim-security-56** (vendor): Cato says AI and security events sit together in one data lake for audit, analytics and response.Quote: "AI and security events sit together for audit, analytics and response."Source: [catonetworks.com](https://www.catonetworks.com/solutions/ai-security/), read 2026-11-16. - **aim-security-57** (none-found): None found on 2026-11-16: the Cato AI security solution page does not name SIEM, SOAR, EDR vendor, identity provider or ticketing integrations for AI events.Source: [catonetworks.com](https://www.catonetworks.com/solutions/ai-security/), read 2026-11-16. - **aim-security-58** (vendor): Cato says agent discovery works by connecting to the customer infrastructure; no deploy time is published for the agents product.Quote: "Cato connects to your infrastructure to discover AI agents across systems, tools, and MCP interactions."Source: [catonetworks.com](https://www.catonetworks.com/platform/agentic-ai-security/), read 2026-11-16. - **aim-security-68** (vendor): Cato says endpoint and network AI activity sit in one dashboard that ties interactions to user, endpoint and data, with blocked and monitored results.Quote: "Endpoint and network AI activity sit in one dashboard, and the same Cato AI engine inspects both."Source: [catonetworks.com](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. - **aim-security-69** (vendor): Cato lists logging violations for auditable evidence and full audit trails for investigation.Quote: "Preserve full audit trails for investigation"Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-end-users/), read 2026-11-16. - **aim-security-70** (vendor): Cato says AI Security for Applications can run on-prem or in the customer cloud with AI Security Outposts so prompts stay in the customer environment.Quote: "Run AI Security for Applications on-prem or in your own cloud with AI Security Outposts. Prompts stay in your environment."Source: [catonetworks.com](https://www.catonetworks.com/solutions/ai-security/), read 2026-11-16. - **aim-security-72** (vendor): Dated news: on 2026-08-03 Cato launched Agentic Threat Prevention, an autonomous-agent capability to predict attack paths; it is network threat prevention, not AI runtime control.Quote: "today announced the launch of Cato Agentic Threat Prevention, a new capability that deploys autonomous agents to predict likely attack paths"Source: [catonetworks.com](https://www.catonetworks.com/news/cato-launches-agentic-threat-prevention-to-predict-mitigate-ai-assisted-attacks/), read 2026-11-16. - **aim-security-73** (vendor): Cato says it can discover agents and MCP servers, then block risky prompts, actions and tool calls.Quote: "Discover agents and MCP servers, then block risky prompts, actions, and tool calls."Source: [catonetworks.com](https://www.catonetworks.com/platform/ai-security-for-end-users/), read 2026-11-16. ## News mentioning Aim Security - [Cato Networks launches Cato AI Endpoint for local agents](/news/cato-ai-endpoint-launch/) ## Related pages - [All vendor profiles](/vendors/) - [Zenity profile](/vendors/zenity/) - [Bay profile](/vendors/bay/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Bay AI agent security URL: https://aiagentsecuritysolutions.com/vendors/bay/ Updated: 2026-11-16 Per Bay's pages, one script pushed through an existing MDM or EDR installs Wave, an ephemeral discovery binary, and Bay Enforcer decides allow, ask, deny or steer on credential access, shell commands and MCP connections before they run, with no proxy. Three third-party reports back the enforcement claims. One policy set spans endpoint, browser and cloud workspaces such as VMs, Kubernetes and Coder, and decision logs export to a SIEM in OCSF format. A gateway layer and named SaaS agent platforms such as Copilot Studio or Agentforce: none found on 16 Nov 2026. Bay is a young company that left stealth in 2026, publishes no price and has little independent testing behind its claims. Bay is a young company that left stealth in 2026, publishes no price and has little independent testing behind its claims. ## Strength and weakness for AI Agent Security Solutions **Strength:** One policy set enforces allow, ask, deny or steer across endpoint, browser and cloud workspaces, with SIEM export (bay-47, bay-37, bay-62). **Weakness:** None found on 2026-11-16 for a gateway layer or named SaaS agent platforms such as Copilot Studio or Agentforce (bay-45, bay-39). ## Scores and evidence Bay is 4th in the overall ranking at 62.0 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Bay: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 8 of 10 | 6 | The vendor covers coding agents, desktop assistants, CLI tools and MCP servers on devices, and third parties describe the on-machine native hook. | [bay-14](#fact-bay-14), [bay-70](#fact-bay-70), [bay-72](#fact-bay-72), [bay-46](#fact-bay-46), [bay-55](#fact-bay-55) | | Coverage: browser layer | 6 of 10 | 6 | The vendor claims browser coverage including browser agents and blocking of domains, extensions and agentic actions; a browser list none found. | [bay-12](#fact-bay-12), [bay-23](#fact-bay-23), [bay-24](#fact-bay-24), [bay-74](#fact-bay-74), [bay-75](#fact-bay-75) | | Coverage: cloud workspace layer | 5 of 10 | 6 | Cloud workspaces such as VMs, Kubernetes and Coder are covered; named SaaS agent platforms none found. | [bay-37](#fact-bay-37), [bay-38](#fact-bay-38), [bay-40](#fact-bay-40), [bay-39](#fact-bay-39) | | Coverage: identity layer | 6 of 10 | 6 | The vendor shows credentials, secrets and permissions each agent can reach and attributes actions; formal revocation and least-privilege workflows are limited. | [bay-41](#fact-bay-41), [bay-42](#fact-bay-42), [bay-43](#fact-bay-43), [bay-68](#fact-bay-68) | | Coverage: gateway layer | 0 of 10 | 6 | None found: the vendor uses an on-device control point and states it uses no proxy. | [bay-45](#fact-bay-45), [bay-44](#fact-bay-44) | | Stops the action | 9 of 10 | 20 | The vendor decides allow, ask, deny or steer on credential access, shell commands and MCP connections before they run, confirmed by three third-party reports. | [bay-47](#fact-bay-47), [bay-48](#fact-bay-48), [bay-53](#fact-bay-53), [bay-54](#fact-bay-54), [bay-55](#fact-bay-55) | | Visibility | 7 of 10 | 15 | An entity graph of agents, MCP servers, SaaS apps and access plus session timelines is described; vendor claims only. | [bay-11](#fact-bay-11), [bay-29](#fact-bay-29), [bay-63](#fact-bay-63), [bay-76](#fact-bay-76) | | Deployment | 8 of 10 | 10 | One MDM or EDR script, no proxy and under five minutes are published and reported by third parties; a supported OS list none found. | [bay-16](#fact-bay-16), [bay-17](#fact-bay-17), [bay-18](#fact-bay-18), [bay-19](#fact-bay-19), [bay-20](#fact-bay-20), [bay-21](#fact-bay-21) | | Policy and governance | 6 of 10 | 10 | IdP-scoped policy, simulation, decision logs and OCSF SIEM export are described; EDR, SOAR and ticketing integrations none found. | [bay-30](#fact-bay-30), [bay-31](#fact-bay-31), [bay-59](#fact-bay-59), [bay-61](#fact-bay-61), [bay-62](#fact-bay-62), [bay-65](#fact-bay-65) | | Transparency | 1 of 10 | 10 | Documentation needs sign-in; a published price and published research none found. | [bay-34](#fact-bay-34), [bay-35](#fact-bay-35), [bay-36](#fact-bay-36) | | Proof | 7 of 10 | 5 | Third parties report $50 million raised and name customers such as Micron and Wix; independent analyst coverage none found. | [bay-06](#fact-bay-06), [bay-07](#fact-bay-07), [bay-08](#fact-bay-08), [bay-09](#fact-bay-09) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 4th | 62.0 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 1st | 71.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 2nd | 65.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | joint 5th | 62.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 4th | 65.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 16th | 47.0 | ## Product, deployment, price and owner - Product assessed: Bay platform: Endpoint, Cloud Workspaces and Browser, with AI Security Posture and Agent Runtime Enforcement - Deployment: Agentless: one script through an existing MDM or EDR runs Wave, an ephemeral binary; Bay Enforcer, a static binary, decides on the device via native hook; no proxy (vendor claims) - Time to deploy: Under 5 minutes, as claimed by the vendor; proof of value in under 14 days (SecurityWeek quote) - Price: none found on 2026-11-16 - Owner: Bay AI, Inc. - Website: [bay.io](https://bay.io/?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor) ## Facts and sources - **bay-01** (vendor): Bay AI, Inc. publishes the Bay site and states its copyright on every page.Quote: "© 2026 Bay AI, Inc. All rights reserved."Source: [bay.io](https://www.bay.io/about?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-02** (vendor): Bay says it was founded by three engineers from endpoint and API security.Quote: "Bay was founded by three engineers who spent their careers solving the hardest problems in endpoint and API security."Source: [bay.io](https://www.bay.io/about?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-03** (third-party): On 2026-11-10 Bay named its leadership: Yuval Zacharia is CEO and co-founder, Aner Morag is CPO and Boris Danilovich is VP of R&D, per SiliconANGLE.Quote: "Zacharia founded Bay with Chief Product Officer Aner Morag and Vice President of Research and Development Boris Danilovich."Source: [siliconangle.com](https://siliconangle.com/2026/11/10/ai-security-startup-bay-launches-with-50m-to-vet-agent-actions-before-they-run/), read 2026-11-16. - **bay-04** (vendor): Bay says it was founded in 2026 and is headquartered in Tel Aviv; its press release is dated 2026-11-10.Quote: "Founded in 2026, Bay is headquartered in Tel Aviv and serves customers in the United States and Europe."Source: [bay.io](https://www.bay.io/blog/bay-emerges-from-stealth-with-50m-after-fortune-100-companies-used-its-platform-to-triple-autonomous-ai-agent-use-2?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-05** (vendor): On 2026-11-10 Bay said it had raised $50 million in total funding.Quote: "Bay has raised $50 million in total funding."Source: [bay.io](https://www.bay.io/blog/bay-emerges-from-stealth-with-50m-after-fortune-100-companies-used-its-platform-to-triple-autonomous-ai-agent-use-2?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-06** (third-party): On 2026-11-11 SecurityWeek reported Bay emerged from stealth with a $37 million Series A led by Battery Ventures plus a $13 million seed.Quote: "emerged from stealth with $50 million funding (comprising a $37 million Series A round led by Battery Ventures"Source: [securityweek.com](https://www.securityweek.com/bay-raises-50-million-to-control-what-enterprise-ai-agents-can-do/), read 2026-11-16. - **bay-07** (third-party): On 2026-11-10 SiliconANGLE reported Micron, Wix, ClickUp and Peloton are among dozens of Bay enterprise customers.Quote: "Micron Technology Inc., Wix.com Ltd., ClickUp Inc. and Peloton Interactive Inc. are among the dozens of enterprise customers signed in the nine months since Bay was founded."Source: [siliconangle.com](https://siliconangle.com/2026/11/10/ai-security-startup-bay-launches-with-50m-to-vet-agent-actions-before-they-run/), read 2026-11-16. - **bay-08** (third-party): The Next Web noted that Bay did not name its Fortune 100 customers or say how it measured the tripling claim, and that one named customer is also an investor.Quote: "The company did not name those customers or say how it measured the figure, and TNW has not verified it."Source: [thenextweb.com](https://thenextweb.com/news/bay-50m-ai-agent-security-stealth-battery-ventures), read 2026-11-16. - **bay-09** (none-found): None found on 2026-11-16: independent analyst coverage of Bay (Gartner, Forrester or similar) on the Bay home page or blog.Source: [bay.io](https://bay.io/?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-10** (vendor): Bay says it is SOC 2 and ISO 27001 certified.Quote: "SOC 2 & ISO 27001 Certified"Source: [bay.io](https://www.bay.io/about?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-11** (vendor): Bay says shadow AI discovery covers agents, models, MCP servers, extensions and skills plus AI built into approved tools, across devices, browsers and cloud workloads.Quote: "Every agent, model, MCP server, extension and skill, plus the AI built into tools your teams already use, across devices, browsers and cloud workloads."Source: [bay.io](https://www.bay.io/shadow-ai-discovery?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-12** (vendor): Bay says its browser coverage finds AI tools, extensions and accounts in every browser.Quote: "Find every AI tool, extension and account in every browser, and keep company data out of personal AI."Source: [bay.io](https://www.bay.io/platform/browser?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-14** (vendor): Bay says its endpoint product finds agents, extensions, CLI tools, MCP servers, skills and models on employee devices.Quote: "Bay finds every agent, extension, CLI tool, MCP server, skill and model on employee devices."Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-16** (vendor): Bay says it deploys with one script through an existing MDM, with no proxy and nothing installed by users.Quote: "Bay deploys with one script through an existing MDM. It does not require a proxy to route traffic through and does not require anything to be installed by users."Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-17** (vendor): Bay says it is agentless and deploys via MDM or EDR using Wave, an ephemeral binary, with no persistent agent and no proxy.Quote: "Bay is agentless: it deploys through your existing MDM or EDR using Wave, an ephemeral binary, so there’s no persistent agent to maintain and no proxy in the path."Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-18** (vendor): Bay says the endpoint inventory appears in under 5 minutes.Quote: "Bay brings it all into view in under 5 minutes and keeps it in bounds, without slowing anyone down."Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-19** (third-party): On 2026-11-10 SiliconANGLE reported Bay installs through the customer's existing MDM or EDR and the company says it takes under five minutes.Quote: "Bay installs through the mobile device management or endpoint detection and response tools a customer already has, and the company says it takes under five minutes."Source: [siliconangle.com](https://siliconangle.com/2026/11/10/ai-security-startup-bay-launches-with-50m-to-vet-agent-actions-before-they-run/), read 2026-11-16. - **bay-20** (third-party): On 2026-11-11 SecurityWeek quoted Bay saying a fleet deploys in under five minutes and a proof of value completes in under 14 days.Quote: "complete a proof of value in less than 14 days."Source: [securityweek.com](https://www.securityweek.com/bay-raises-50-million-to-control-what-enterprise-ai-agents-can-do/), read 2026-11-16. - **bay-21** (none-found): None found on 2026-11-16: a list of supported operating systems for the Bay endpoint scanner, on the endpoint product page and the platform page.Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-22** (vendor): Bay says it identifies personal versus corporate AI accounts including in personal Chrome profiles; no full browser list was found.Quote: "including in personal Chrome profiles"Source: [bay.io](https://www.bay.io/platform/browser?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-23** (vendor): Bay says browser guardrails keep personal data, secrets and source code out of prompts, with allow, warn, steer or block per team.Quote: "Keep personal data, secrets and source code out of prompts, and decide what to, Allow, Warn, Steer or Block, tailored to each team."Source: [bay.io](https://www.bay.io/platform/browser?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-24** (vendor): Bay says it can allow or block specific AI domains and extensions and stop risky agentic actions inside the browser.Quote: "Bay lets you allow or block specific AI domains and extensions, and stops risky agentic actions inside the browser"Source: [bay.io](https://www.bay.io/platform/browser?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-26** (vendor): Bay says every plugin, package, skill and MCP server is scanned and vetted before approval and each time it changes.Quote: "Every plugin, package, skill and MCP server is scanned and vetted before it’s approved and every time it changes."Source: [bay.io](https://www.bay.io/shadow-ai-discovery?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-29** (vendor): Bay says it builds an entity graph of the user behind each agent, the MCP servers and SaaS apps it touches, and the access it holds.Quote: "Bay builds a contextual entity graph of how everything connects: the user behind each agent, the MCP servers and SaaS apps it touches, and the access it holds."Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-30** (vendor): Bay says it connects to the identity provider so policies apply to the right people and groups.Quote: "Bay connects to your identity provider, so every policy applies to the right people and groups, and keeps up as your organization changes."Source: [bay.io](https://www.bay.io/ai-governance-compliance?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-31** (vendor): Bay says policies can be scoped by IdP group, device or runtime and can set which MCP servers may read or write.Quote: "Policies can be scoped by IdP group, device or runtime, and can set which MCP servers can read or write"Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-34** (observed): Bay documentation at docs.bay.io redirects to a sign-in page and is open only to signed-in Bay users.Quote: "The Bay documentation is open only to signed-in Bay users."Source: [docs.bay.io](https://docs.bay.io/?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-35** (none-found): None found on 2026-11-16: a published price. The site has no pricing page (bay.io/pricing returns 404) and the demo page only offers a 30-minute demo.Source: [bay.io](https://www.bay.io/demo?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-36** (none-found): None found on 2026-11-16: published research, threat reports or open-source code; the Bay blog lists only two announcement posts dated 2026-11-10.Source: [bay.io](https://www.bay.io/blog?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-37** (vendor): Bay says it discovers agents across virtual machines, Kubernetes, Coder workspaces and hosted sessions, with the MCP servers and skills each uses.Quote: "Discover every agent across virtual machines, Kubernetes, Coder workspaces and hosted sessions, along with the MCP servers and skills each one uses."Source: [bay.io](https://www.bay.io/platform/cloud-workspaces?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-38** (vendor): Bay says teams run Claude Code, Codex, Gemini CLI and GitHub Copilot in cloud workspaces and that Bay covers them.Quote: "Teams now run agents like Claude Code, Codex, Gemini CLI and GitHub Copilot in clusters, cloud workspaces and hosted sessions"Source: [bay.io](https://www.bay.io/platform/cloud-workspaces?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-39** (none-found): None found on 2026-11-16: named support for SaaS agent platforms (Microsoft Copilot Studio, Salesforce Agentforce, ServiceNow, AWS Bedrock agents) on the cloud workspaces page.Source: [bay.io](https://www.bay.io/platform/cloud-workspaces?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-40** (third-party): On 2026-11-10 SiliconANGLE reported Bay tracks agents on a laptop, in a cloud workload or in a SaaS application under one policy set.Quote: "whether they sit on an employee’s laptop, inside a cloud workload or in a software-as-a-service application"Source: [siliconangle.com](https://siliconangle.com/2026/11/10/ai-security-startup-bay-launches-with-50m-to-vet-agent-actions-before-they-run/), read 2026-11-16. - **bay-41** (vendor): Bay says it shows which credentials, secrets and permissions each cloud agent can reach so risky ones can be locked down.Quote: "See which credentials, secrets and permissions each cloud agent can reach, and lock down the risky ones before a long-running task puts them to use."Source: [bay.io](https://www.bay.io/platform/cloud-workspaces?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-42** (vendor): Bay says it finds credentials, tokens and secrets that agents and MCP servers can read and helps lock them down.Quote: "Bay finds the credentials, tokens and secrets that agents and MCP servers can read, and helps you lock them down before a long-running task puts them to use."Source: [bay.io](https://www.bay.io/ai-security-posture?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-43** (vendor): Bay says it attributes each action to the person or the agent acting for them and records the agent's permission mode.Quote: "Bay attributes each action to the person who took it or the agent that acted for them, and records the permission mode the agent was running in."Source: [bay.io](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-44** (vendor): Bay says it uses no proxy; decisions are made on the device.Quote: "No proxy: every decision is made on the device."Source: [bay.io](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-45** (none-found): None found on 2026-11-16: a gateway or proxy product between agents and models, tools or MCP servers; Bay pages describe an on-device control point instead.Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-46** (third-party): On 2026-11-10 SiliconANGLE reported the AI tool hands each proposed action to Bay Enforcer, a static binary, through a native hook on the machine.Quote: "Each time an agent proposes an action, the AI tool running it hands the request to Bay Enforcer, a static binary, through a native hook."Source: [siliconangle.com](https://siliconangle.com/2026/11/10/ai-security-startup-bay-launches-with-50m-to-vet-agent-actions-before-they-run/), read 2026-11-16. - **bay-47** (vendor): Bay says runtime enforcement decides allow, ask, deny or steer on every action, locally and in milliseconds with no cloud round-trip.Quote: "Bay reads each agent’s whole session to understand its intent, then decides on every action: allow, ask, deny or steer."Source: [bay.io](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-48** (vendor): Bay says it acts on credential access, risky shell commands and untrusted MCP server connections in milliseconds.Quote: "When an agent reaches for credentials, runs a risky shell command or connects to an untrusted MCP server, Bay allows, asks, denies or steers the action in milliseconds"Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-49** (vendor): Bay says endpoint decisions are made on the device in under 4ms.Quote: "allowing, asking, denying or steering it right on the device in under 4ms"Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-51** (vendor): Bay says it can prevent prompt injection attempts at runtime.Quote: "Bay can also prevent prompt injection attempts at runtime"Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-53** (third-party): On 2026-11-10 SiliconANGLE reported that anything Bay judges too risky never runs.Quote: "Anything judged too risky never runs."Source: [siliconangle.com](https://siliconangle.com/2026/11/10/ai-security-startup-bay-launches-with-50m-to-vet-agent-actions-before-they-run/), read 2026-11-16. - **bay-54** (third-party): On 2026-11-11 SecurityWeek reported Bay decides in milliseconds to allow, refer to the user, redirect or deny each action.Quote: "decides in milliseconds whether to allow the action, refer it to the user, redirect it based on policy, or simply deny it"Source: [securityweek.com](https://www.securityweek.com/bay-raises-50-million-to-control-what-enterprise-ai-agents-can-do/), read 2026-11-16. - **bay-55** (third-party): On 2026-11-10 The Next Web reported Bay checks every action an agent proposes before it runs, then allows, blocks or asks a person.Quote: "It checks every action an agent proposes before it runs, then allows it, blocks it or asks a person."Source: [thenextweb.com](https://thenextweb.com/news/bay-50m-ai-agent-security-stealth-battery-ventures), read 2026-11-16. - **bay-58** (vendor): Bay says it applies intent-aware enforcement on every tool call in cloud agents.Quote: "Intent-aware enforcement on every tool call."Source: [bay.io](https://www.bay.io/platform/cloud-workspaces?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-59** (vendor): Bay says any policy can run in simulation to show what it would allow, ask, deny or steer before users are affected.Quote: "Run any policy in simulation to see what each rule would allow, ask, deny or steer, before a single user is affected."Source: [bay.io](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-60** (vendor): Bay says it replays 30 days of agent history to show what a policy would do before enforcement.Quote: "Before enforcement, Bay replays 30 days of agent history to show what a policy would do before any user is affected."Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-61** (vendor): Bay says every decision is logged against the policy behind it for auditors.Quote: "Every decision is logged against the policy behind it, ready for your auditors."Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-62** (vendor): Bay says events reach the customer SIEM in OCSF format.Quote: "Events arrive in your SIEM in OCSF format, with enough context that analysts can triage without opening another console."Source: [bay.io](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-63** (vendor): Bay says it records every prompt, tool call and system action on one timeline per session.Quote: "Every prompt, tool call and system action on one timeline, so you can see how an agent got from its first instruction to its last action."Source: [bay.io](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-64** (vendor): Bay says it maps policies and activity to OWASP LLM Top 10, NIST AI RMF, MITRE ATLAS, the EU AI Act and ISO/IEC 42001.Quote: "Bay links your policies and the activity behind them to the OWASP LLM Top 10, NIST AI RMF, MITRE ATLAS, the EU AI Act and ISO/IEC 42001"Source: [bay.io](https://www.bay.io/ai-governance-compliance?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-65** (none-found): None found on 2026-11-16: named EDR, SOAR or ticketing integrations beyond SIEM export, on the investigation page.Source: [bay.io](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-66** (vendor): Bay says it flags risky configurations such as bypass-permissions mode and sandboxes switched off, rates severity and shows the fix.Quote: "Bypass-permissions mode, sandboxes switched off, unrestricted tool access. Bay spots each one, rates how serious it is and shows you how to fix it."Source: [bay.io](https://www.bay.io/ai-security-posture?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-67** (vendor): Bay says it finds toxic combinations such as an agent that can read secrets and also reach the internet.Quote: "Some risks only appear together, like an agent that can read secrets and also reach the internet."Source: [bay.io](https://www.bay.io/ai-security-posture?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-68** (vendor): Bay's product walkthrough says a discovered Claude Code install was connected to Salesforce, Atlassian and Sentry and could read plaintext secrets in its MCP configs.Quote: "the agent can read plaintext secrets in its MCP configs, as well as AWS and SSH keys on disk"Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-70** (vendor): Bay names Claude Code and Cursor as coding agents it secures on the endpoint.Quote: "How do you secure AI coding agents like Claude Code and Cursor on the endpoint?"Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-72** (vendor): Bay describes its endpoint scope as coding agents, desktop assistants, MCP servers and skills on employee devices.Quote: "Coding agents, desktop assistants, MCP servers and skills, working with access to code, data and production environments."Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-73** (vendor): Bay says it follows each coding agent session from prompt to tool call to system action.Quote: "Bay secures AI coding agents by following each session from prompt to tool call to system action."Source: [bay.io](https://www.bay.io/platform/ai-endpoint-security?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-74** (vendor): Bay says it provides AI browser security across the browsers a team already uses and covers browsers that act on their own.Quote: "Bay provides AI browser security across the browsers your teams already use."Source: [bay.io](https://www.bay.io/platform/browser?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-75** (vendor): Bay says browser agents that click and type on their own are in scope.Quote: "and now browsers that click and type on their own"Source: [bay.io](https://www.bay.io/platform/browser?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-76** (vendor): Bay says its records include full transcripts of prompts, actions and MCP calls traceable to a person.Quote: "Full transcripts of prompts, actions and MCP calls - traceable to the person behind each session."Source: [bay.io](https://www.bay.io/platform/cloud-workspaces?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. - **bay-77** (none-found): None found on 2026-11-16: a per-server allowlist or registry for MCP servers beyond scoping which servers may read or write; see approval workflow mention.Source: [bay.io](https://www.bay.io/platform?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=vendor), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Aim Security profile](/vendors/aim-security/) - [Check Point profile](/vendors/check-point/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Check Point AI agent security URL: https://aiagentsecuritysolutions.com/vendors/check-point/ Updated: 2026-11-16 Check Point documents agent discovery for Amazon Bedrock, Bedrock AgentCore, Google Cloud, Microsoft Copilot Studio and Salesforce Agentforce, recording each agent's owner, model, tools, MCP servers and recent activity in a shared inventory. It says AI Agent Security assesses risk per agent and evaluates tool calls in context, with allow and deny lists and a beta deviation detector. A Guard API screens prompts, tool calls and responses, and AI Guardrails deploy as SaaS, Helm, Docker or air-gapped. Documentation and an OpenAPI spec are public. Agent identity, token and revocation controls: none found on 16 Nov 2026, and native platform runtime integrations are roadmap only. ## Strength and weakness for AI Agent Security Solutions **Strength:** Agent discovery across seven named platforms with per-agent risk ratings plus runtime tool-call blocking through one Guard API (check-point-15, check-point-21, check-point-25). **Weakness:** No agent identity, token or revocation controls found on 2026-11-16, and native platform runtime integrations are roadmap only (check-point-19, check-point-17). ## Scores and evidence Check Point is 12th in the overall ranking at 56.2 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Check Point: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 4 of 10 | 6 | An optional desktop agent covers code assistants and desktop agents; OS list, local MCP controls and CLI coverage none found. | [check-point-08](#fact-check-point-08), [check-point-09](#fact-check-point-09), [check-point-10](#fact-check-point-10), [check-point-11](#fact-check-point-11) | | Coverage: browser layer | 5 of 10 | 6 | A browser extension redacts prompts and restricts paste for public AI tools; browser agent control none found. | [check-point-12](#fact-check-point-12), [check-point-13](#fact-check-point-13), [check-point-14](#fact-check-point-14) | | Coverage: cloud workspace layer | 6 of 10 | 6 | Discovery for Bedrock, AgentCore, Copilot Studio and Agentforce is documented; native runtime integrations are roadmap and the product is early access. | [check-point-15](#fact-check-point-15), [check-point-16](#fact-check-point-16), [check-point-17](#fact-check-point-17), [check-point-18](#fact-check-point-18) | | Coverage: identity layer | 1 of 10 | 6 | Per-agent risk across tools is assessed, but the docs defer identity to other products; credential inventory and revocation none found. | [check-point-19](#fact-check-point-19), [check-point-20](#fact-check-point-20), [check-point-21](#fact-check-point-21) | | Coverage: gateway layer | 6 of 10 | 6 | The Guard API screens prompts, tool calls and responses inline; a dedicated gateway is only a planned partnership. | [check-point-22](#fact-check-point-22), [check-point-23](#fact-check-point-23), [check-point-24](#fact-check-point-24) | | Stops the action | 6 of 10 | 20 | Tool allow and deny lists and a beta deviation detector act at runtime; the pre-execution monitor is research with unstated status. | [check-point-25](#fact-check-point-25), [check-point-26](#fact-check-point-26), [check-point-27](#fact-check-point-27), [check-point-28](#fact-check-point-28) | | Visibility | 7 of 10 | 15 | Discovery records each agent's owner, model, tools, MCP servers and recent activity in a shared inventory; vendor documentation only. | [check-point-15](#fact-check-point-15), [check-point-16](#fact-check-point-16), [check-point-21](#fact-check-point-21), [check-point-29](#fact-check-point-29) | | Deployment | 6 of 10 | 10 | SaaS, Helm, Docker, air-gapped and extension options are documented; a time to deploy for agent security none found. | [check-point-10](#fact-check-point-10), [check-point-12](#fact-check-point-12), [check-point-30](#fact-check-point-30), [check-point-31](#fact-check-point-31), [check-point-32](#fact-check-point-32) | | Policy and governance | 4 of 10 | 10 | Unified audit trails and a policy simulator exist; SIEM, EDR, IdP and ticketing integrations none found. | [check-point-27](#fact-check-point-27), [check-point-33](#fact-check-point-33), [check-point-34](#fact-check-point-34), [check-point-35](#fact-check-point-35) | | Transparency | 7 of 10 | 10 | Documentation, an OpenAPI spec and research are public; a published price none found. | [check-point-37](#fact-check-point-37), [check-point-39](#fact-check-point-39), [check-point-38](#fact-check-point-38) | | Proof | 7 of 10 | 5 | Third parties report the acquisition and launch; named customers include Dropbox, Nubank and Clalit. | [check-point-02](#fact-check-point-02), [check-point-03](#fact-check-point-03), [check-point-40](#fact-check-point-40), [check-point-41](#fact-check-point-41) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 12th | 56.2 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 13th | 55.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 8th | 58.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 8th | 61.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 14th | 46.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 9th | 61.0 | ## Product, deployment, price and owner - Product assessed: Check Point AI Agent Security (discovery, risk assessment, runtime via AI Guardrails) plus Workforce AI Security for endpoint and browser coverage - Deployment: Managed SaaS or self-hosted (Kubernetes Helm, Docker, air-gapped) for AI Guardrails via API; browser extension plus optional desktop agent for Workforce AI Security; agent platform connectors for discovery - Time to deploy: Workforce AI Security says it deploys instantly through a browser extension; none found on 2026-11-16 for AI Agent Security and AI Guardrails - Price: none found on 2026-11-16 - Owner: Check Point Software Technologies Ltd. - Website: checkpoint.com ## Facts and sources - **check-point-01** (vendor): Check Point Software Technologies announced on 2025-09-16 that it agreed to acquire Lakera, an AI-native security platform; the press release says Lakera products became Check Point AI Red Teaming and Check Point AI Agent Security.Quote: "Its flagship solutions, Check Point AI Red Teaming and Check Point AI Agent Security, provide pre-deployment posture assessments and real-time runtime enforcement"Source: [checkpoint.com](https://www.checkpoint.com/press-releases/check-point-acquires-lakera-to-deliver-end-to-end-ai-security-for-enterprises/), read 2026-11-16. - **check-point-02** (third-party): SecurityWeek reported on 2025-09-16 that Check Point planned to acquire Lakera, and that on closing Lakera would form the basis of a Global Center of Excellence for AI Security.Quote: "Upon closing, Lakera will form the basis of Check Point’s new Global Center of Excellence for AI Security"Source: [securityweek.com](https://www.securityweek.com/check-point-to-acquire-ai-security-firm-lakera/), read 2026-11-16. - **check-point-03** (third-party): Help Net Security reported on 2026-03-24 that Check Point launched the AI Defense Plane, built on the Lakera and Cyata acquisitions, with workforce, AI application and agent, and red teaming modules.Quote: "strengthened by technologies from ThreatCloud AI and the company’s recent Lakera and Cyata acquisitions"Source: [helpnetsecurity.com](https://www.helpnetsecurity.com/2026/03/24/check-point-ai-defense-plane/), read 2026-11-16. - **check-point-04** (vendor): Check Point lists Workforce AI Security, AI Red Teaming and AI Agent Security as its AI Security products, and describes an AI Defense Plane to discover, protect and govern AI.Quote: "Check Point delivers one platform to discover, protect, and govern every AI interaction from employees to agents."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/), read 2026-11-16. - **check-point-05** (vendor): The Lakera documentation site is now branded Check Point AI Security, with products AI Agent Security, AI Guardrails and AI Red Teaming.Quote: "AI Guardrails is the runtime layer of AI Agent Security, and is also sold on its own"Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security), read 2026-11-16. - **check-point-06** (vendor): The Lakera website still operates as lakera.ai and describes workforce AI security, AI agent security and AI red teaming products.Quote: "Shadow AI discovery across apps and browsers"Source: [lakera.ai](https://www.lakera.ai), read 2026-11-16. - **check-point-07** (vendor): A 2026-10-01 Check Point press release names Adam Ely as General Manager, AI Security at Check Point.Quote: "said Adam Ely, General Manager, AI Security at Check Point Software Technologies"Source: [checkpoint.com](https://www.checkpoint.com/press-releases/clalit-health-services-and-check-point-announce-strategic-partnership-to-secure-ai-adoption-across-healthcare-environments/), read 2026-11-16. - **check-point-08** (vendor): Check Point says Workforce AI Security covers desktop agents, code assistants such as GitHub Copilot and Cursor, and MCP-connected workflows; the Enterprise tier covers web, desktop, MSP and IDEs.Quote: "The Enterprise tier covers web, desktop, MSP, and IDEs, while the Essentials tier focuses on web applications."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/workforce-ai-security/), read 2026-11-16. - **check-point-09** (vendor): Check Point says Workforce AI Security discovers AI tools across web apps, desktop agents, browser extensions, SaaS integrations, code assistants and MCPs, with an application catalog and risk scoring.Quote: "automatically discovers every AI tool in use, sanctioned or unsanctioned across web apps, desktop agents, browser extensions, SaaS integrations, code assistants, and MCPs"Source: [checkpoint.com](https://www.checkpoint.com/ai-security/workforce-ai-security/), read 2026-11-16. - **check-point-10** (vendor): Check Point says endpoint coverage comes from an optional desktop agent, while the base deployment is a browser extension.Quote: "Desktop agent deployment is also available for broader coverage across device-level AI usage."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/workforce-ai-security/), read 2026-11-16. - **check-point-11** (none-found): None found on 2026-11-16: looked for named operating systems, local MCP server controls and command-line agent coverage for the desktop agent on the Workforce AI Security page and the docs index.Source: [checkpoint.com](https://www.checkpoint.com/ai-security/workforce-ai-security/), read 2026-11-16. - **check-point-12** (vendor): Check Point says Workforce AI Security deploys through a lightweight browser extension with no architecture changes.Quote: "Workforce AI Security deploys instantly through a lightweight browser extension, no complex setup, no downtime, and no architecture changes required."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/workforce-ai-security/), read 2026-11-16. - **check-point-13** (vendor): Check Point says Workforce AI Security protects data in prompts with inline redaction, copy and paste restrictions, file and image redaction with OCR, and user guidance.Quote: "Protections include inline redaction, copy/paste restrictions, file and image redaction with OCR, and interactive user guidance."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/workforce-ai-security/), read 2026-11-16. - **check-point-14** (vendor): Check Point says Workforce AI Security covers public AI tools, browser extensions and SaaS-integrated AI features.Quote: "Workforce AI Security covers the full spectrum of employee AI interactions: public AI tools (ChatGPT, Gemini, Claude, Copilot), browser extensions, desktop agents"Source: [checkpoint.com](https://www.checkpoint.com/ai-security/workforce-ai-security/), read 2026-11-16. - **check-point-15** (vendor): Check Point documents agent discovery for Amazon Bedrock, Bedrock AgentCore, Google Cloud, Microsoft Copilot Studio, Salesforce Agentforce, n8n and Relevance AI.Quote: "Discovery is available today for: Amazon Bedrock Amazon Bedrock AgentCore Google Cloud Microsoft Copilot Studio Salesforce Agentforce n8n Relevance AI"Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security/discovery), read 2026-11-16. - **check-point-16** (vendor): Check Point documents that discovery records each agent's name, owner, model, tools, connected MCP servers and recent activity, with depth varying by platform.Quote: "Connected MCP servers and the tools they provide."Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security/discovery), read 2026-11-16. - **check-point-17** (vendor): Check Point documents that native platform runtime integrations are on the roadmap, so runtime protection for cloud-hosted agents goes through the Guard API today.Quote: "native platform runtime integrations are on the roadmap"Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security), read 2026-11-16. - **check-point-18** (vendor): Check Point states that AI Agent Security entered Early Access on 2026-06-01, adding discovery and risk assessment to AI Guardrails.Quote: "Check Point AI Agent Security extends AI Guardrails with discovery and risk assessment for the agents your organization builds and deploys"Source: [docs.lakera.ai](https://docs.lakera.ai/changelog), read 2026-11-16. - **check-point-19** (none-found): None found on 2026-11-16: looked for agent identity inventory, token or credential management and permission revocation on the AI Agent Security page and docs; the docs defer identity to other products.Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security), read 2026-11-16. - **check-point-20** (vendor): Check Point docs say identity, cloud and network controls sit outside AI Agent Security and come from other parts of the architecture.Quote: "controls elsewhere in your architecture, such as your identity provider, cloud security, and network controls"Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security), read 2026-11-16. - **check-point-21** (vendor): Check Point says AI Agent Security assesses agent risk across tools, skills, MCP servers, usage and connected components, and the docs add per-agent risk ratings.Quote: "Understand agent risk across tools, skills, MCP servers, usage, and connected components."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/ai-agent-security/), read 2026-11-16. - **check-point-22** (vendor): Check Point documents the Guard API as the single integration point that screens prompts, model outputs, tool calls, tool responses and tool descriptions for an agent workflow.Quote: "applied across the agentic workflow via the Guard API: user prompts, model outputs, tool calls, tool responses, and tool descriptions"Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security), read 2026-11-16. - **check-point-23** (vendor): Check Point says AI Agent Security acts as an inline enforcement layer that inspects prompts, responses, external content and agent tool calls before risky actions run.Quote: "AI Agent Security acts as an inline enforcement layer for AI agent interactions."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/ai-agent-security/), read 2026-11-16. - **check-point-24** (vendor): On 2026-10-01 Check Point and Clalit Health Services announced a partnership to jointly build a dedicated AI Gateway between users, applications and AI models.Quote: "Check Point and Clalit development teams will jointly build a dedicated control and security layer between users, applications, and AI models."Source: [checkpoint.com](https://www.checkpoint.com/press-releases/clalit-health-services-and-check-point-announce-strategic-partnership-to-secure-ai-adoption-across-healthcare-environments/), read 2026-11-16. - **check-point-25** (vendor): Check Point says AI Agent Security evaluates agent tool calls in context to help block unsafe or unauthorized behavior at runtime.Quote: "Evaluate agent tool calls and actions in context, helping block unsafe or unauthorized behavior at runtime."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/ai-agent-security/), read 2026-11-16. - **check-point-26** (vendor): Check Point docs say the Dangerous Deviation detector (beta) flags agent tool calls that cause data leakage, privilege escalation or system destruction with no warrant in the trusted conversation.Quote: "Access and privilege escalation : public sharing of resources, sudo or admin grants, widened ACLs, or attempts to escalate permissions beyond a sandbox."Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-behavior-defense), read 2026-11-16. - **check-point-27** (vendor): Check Point docs describe a Tool Allow/Deny List enforced through Guard API screening, and a project Detect mode that monitors without blocking.Quote: "whether the tools it calls are allowed at all"Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-behavior-defense), read 2026-11-16. - **check-point-28** (vendor): A Check Point research blog dated 2026-09-15 describes a sidecar control monitor that prevents harmful agent actions before they execute at sub-100ms latency; the post is research, and its product status is not stated there.Quote: "prevents harmful actions before they execute in real time at sub-100ms latency"Source: [blog.checkpoint.com](https://blog.checkpoint.com/ai-security/synchronous-control-monitoring-preventing-harmful-agent-actions-in-real-time/), read 2026-11-16. - **check-point-29** (vendor): Check Point says the agent inventory is shared between AI Agent Security and Workforce AI Security, so on-device agents appear alongside platform agents.Quote: "agents discovered on employee devices appear through Workforce AI Security"Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security/discovery), read 2026-11-16. - **check-point-30** (vendor): Check Point documents AI Guardrails as managed SaaS or self-hosted on Kubernetes (Helm), Docker, or air-gapped.Quote: "AI Guardrails is available as a fully managed SaaS solution or as a self-hosted deployment ."Source: [docs.lakera.ai](https://docs.lakera.ai/docs/selfhosting), read 2026-11-16. - **check-point-31** (vendor): Check Point documents that runtime protection integrates through the Guard API, with discovery and risk assessment managed in the portal.Quote: "Discovery and risk assessment are posture capabilities, managed in the portal."Source: [docs.lakera.ai](https://docs.lakera.ai/docs/agent-security), read 2026-11-16. - **check-point-32** (none-found): None found on 2026-11-16: looked for a published time to deploy for AI Agent Security and AI Guardrails in the docs quickstart and product pages; only Workforce AI claims "instantly".Source: [docs.lakera.ai](https://docs.lakera.ai/docs/api/guard), read 2026-11-16. - **check-point-33** (vendor): Check Point says the AI Defense Plane applies consistent policy with unified audit trails and compliance reporting.Quote: "Apply consistent policy across all AI interactions, with unified audit trails and compliance reporting."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/), read 2026-11-16. - **check-point-34** (vendor): Check Point added a Policy Impact Simulator on 2026-04-10 that shows how sensitivity levels would have affected historical traffic.Quote: "an interactive tool that shows how different sensitivity levels and guardrail configurations would have affected your historical traffic"Source: [docs.lakera.ai](https://docs.lakera.ai/changelog), read 2026-11-16. - **check-point-35** (none-found): None found on 2026-11-16: looked for named SIEM, EDR, identity provider and ticketing integrations for AI Agent Security and AI Guardrails on the product pages, integration guide and docs index.Source: [docs.lakera.ai](https://docs.lakera.ai/docs/integration), read 2026-11-16. - **check-point-36** (vendor): Check Point docs provide a request latency percentiles chart on the Analytics page (2026-10-26) and a Guard logs API.Quote: "Request latency percentiles chart on the Analytics page"Source: [docs.lakera.ai](https://docs.lakera.ai/changelog), read 2026-11-16. - **check-point-37** (vendor): Check Point publishes AI Security documentation, an OpenAPI spec and a changelog that are readable without sign-in.Quote: "Subscribe via RSS"Source: [docs.lakera.ai](https://docs.lakera.ai/changelog), read 2026-11-16. - **check-point-38** (none-found): None found on 2026-11-16: looked for a published price for AI Agent Security, Workforce AI Security and AI Guardrails on the Check Point AI Security pages and the Lakera site; the pages only offer demos.Source: [checkpoint.com](https://www.checkpoint.com/ai-security/), read 2026-11-16. - **check-point-39** (vendor): Check Point publishes AI agent security research on its blog, including a post dated 2026-09-15 on synchronous control monitoring.Quote: "Control monitors ingest the trace as context."Source: [blog.checkpoint.com](https://blog.checkpoint.com/ai-security/synchronous-control-monitoring-preventing-harmful-agent-actions-in-real-time/), read 2026-11-16. - **check-point-40** (vendor): The AI Agent Security page quotes a Dropbox security engineer and the Nubank CISO as customers of the Lakera technology.Quote: "The Lakera team has accelerated our GenAI journey, allowing us to create secure GenAI experiences at scale."Source: [checkpoint.com](https://www.checkpoint.com/ai-security/ai-agent-security/), read 2026-11-16. - **check-point-41** (vendor): On 2026-10-01 Check Point announced Clalit Health Services will deploy Check Point AI Security for real-time visibility and control over AI usage and agents.Quote: "Check Point AI Security will give Clalit real-time visibility and control over AI usage and AI agents across the organization."Source: [checkpoint.com](https://www.checkpoint.com/press-releases/clalit-health-services-and-check-point-announce-strategic-partnership-to-secure-ai-adoption-across-healthcare-environments/), read 2026-11-16. - **check-point-52** (vendor): On 2026-07-30 Check Point announced a new AI Network Firewall, listed in its newsroom.The exact quote is on the source page.Source: [checkpoint.com](https://www.checkpoint.com/press-releases/), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Bay profile](/vendors/bay/) - [Cisco profile](/vendors/cisco/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Cisco AI agent security URL: https://aiagentsecuritysolutions.com/vendors/cisco/ Updated: 2026-11-16 Cisco combines open-source and commercial pieces. The DefenseClaw documentation says its guardrail checks a prompt, a shell command or a tool call before the agent acts, with fail-closed defaults, and marks support on Linux, macOS and Windows only after a live test. The AI Defense plugin for Kong Gateway inspects LLM and MCP requests, and audit logs export to Splunk or OpenTelemetry. Code is Apache-2.0 and an Explorer Edition is described as free. Cisco says agent identity and its MCP gateway are still in development, and a central agent inventory is noted as in development. Browser agents and Copilot Studio, Agentforce or ServiceNow coverage: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers endpoint hooks on many named coding agents on three operating systems plus a gateway plugin and audit export to Splunk or OpenTelemetry (cisco-10, cisco-11, cisco-05, cisco-17). **Weakness:** None found on 2026-11-16 for browser agents or named SaaS agent platforms, and Cisco says agent identity and MCP gateway features are still in development (cisco-46, cisco-47, cisco-41). ## Scores and evidence Cisco is 10th in the overall ranking at 57.0 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Cisco: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 7 of 10 | 6 | Documented hooks for named coding agents block tool calls and shell commands, live-tested per OS; vendor documentation only. | [cisco-10](#fact-cisco-10), [cisco-11](#fact-cisco-11), [cisco-12](#fact-cisco-12), [cisco-14](#fact-cisco-14) | | Coverage: browser layer | 1 of 10 | 6 | Only network-based shadow AI detection; browser agent coverage none found. | [cisco-47](#fact-cisco-47), [cisco-48](#fact-cisco-48) | | Coverage: cloud workspace layer | 5 of 10 | 6 | AgentCore, Strands and Google ADK inspection exist, the ADK plugin observed; Copilot Studio, Agentforce and ServiceNow none found. | [cisco-43](#fact-cisco-43), [cisco-44](#fact-cisco-44), [cisco-45](#fact-cisco-45), [cisco-46](#fact-cisco-46) | | Coverage: identity layer | 6 of 10 | 6 | Placeholder credentials and task-scoped privileged access exist; agent registration is reported by third parties as planned. | [cisco-15](#fact-cisco-15), [cisco-39](#fact-cisco-39), [cisco-40](#fact-cisco-40), [cisco-55](#fact-cisco-55) | | Coverage: gateway layer | 6 of 10 | 6 | A Kong plugin inspects LLM and MCP traffic and fails closed; the MCP gateway is in development. | [cisco-05](#fact-cisco-05), [cisco-07](#fact-cisco-07), [cisco-41](#fact-cisco-41) | | Stops the action | 7 of 10 | 20 | Prompts, shell commands and tool calls are checked before acting with fail-closed defaults; the docs state a limit on child processes. | [cisco-02](#fact-cisco-02), [cisco-05](#fact-cisco-05), [cisco-09](#fact-cisco-09), [cisco-10](#fact-cisco-10), [cisco-20](#fact-cisco-20) | | Visibility | 5 of 10 | 15 | Decision logs exist, but the central agent inventory is noted as still in development. | [cisco-17](#fact-cisco-17), [cisco-42](#fact-cisco-42), [cisco-43](#fact-cisco-43) | | Deployment | 6 of 10 | 10 | Network enforcement and package or MDM rollout are documented; a time to deploy for the runtime product none found. | [cisco-04](#fact-cisco-04), [cisco-38](#fact-cisco-38) | | Policy and governance | 6 of 10 | 10 | Group-based policy and OpenTelemetry and webhook log export are documented; IdP and ticketing integrations none found. | [cisco-16](#fact-cisco-16), [cisco-17](#fact-cisco-17), [cisco-18](#fact-cisco-18), [cisco-53](#fact-cisco-53) | | Transparency | 6 of 10 | 10 | Documentation and Apache-2.0 code are public with a free edition; a paid list price none found. | [cisco-29](#fact-cisco-29), [cisco-56](#fact-cisco-56), [cisco-57](#fact-cisco-57) | | Proof | 5 of 10 | 5 | Third-party coverage and one named customer voice; analyst coverage none found for this line. | [cisco-58](#fact-cisco-58), [cisco-60](#fact-cisco-60), [cisco-59](#fact-cisco-59) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 10th | 57.0 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 7th | 63.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 12th | 45.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 10th | 57.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 7th | 60.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 10th | 60.0 | ## Product, deployment, price and owner - Product assessed: Cisco AI Defense (runtime guardrails, AI Cloud Visibility, Explorer red teaming) with open-source DefenseClaw and Duo IAM agent registration (in development) - Deployment: Network-level enforcement without agents or libraries for AI Defense; Kong plugin, Python SDK, command-line MCP Scanner, and DefenseClaw hooks installed per user or by MDM package - Time to deploy: Cisco says DefenseClaw installs in about five minutes (SiliconANGLE); none found on 2026-11-16 for AI Defense - Price: none found on 2026-11-16 (paid AI Defense); Explorer Edition is described as free - Owner: Cisco Systems, Inc. - Website: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html) ## Facts and sources - **cisco-01** (vendor): Cisco Systems, Inc. is the owner and copyright holder of DefenseClaw, the open-source agent governance layer published in the cisco-ai-defense GitHub organization alongside AI Defense.The exact quote is on the source page.Source: [github.com](https://github.com/cisco-ai-defense/defenseclaw), read 2026-11-16. - **cisco-02** (vendor): Cisco says AI Defense runtime protection puts guardrails in the network and blocks adversarial attacks and harmful responses in real time.Quote: "Protect production AI applications with guardrails embedded in the network. Block adversarial attacks and harmful responses in real time."Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-04** (vendor): Cisco says AI Defense enforces at the network level, with no agents or libraries needed on the application side.Quote: "Cisco enforces AI security at the network-level without the need for agents or libraries."Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-05** (vendor): The AI Defense Intercept plugin for Kong Gateway inspects AI/LLM and MCP requests and responses and blocks or forwards traffic on the AI Defense verdict.Quote: "It blocks or forwards traffic based on the AI Defense verdict."Source: [github.com](https://github.com/cisco-ai-defense/ai-defense-kong-plugin), read 2026-11-16. - **cisco-07** (vendor): The Kong plugin fails closed by default: traffic is blocked when inspection is unavailable, and an operator can set it to fail open.Quote: "Block traffic when inspection is unavailable."Source: [github.com](https://github.com/cisco-ai-defense/ai-defense-kong-plugin), read 2026-11-16. - **cisco-09** (vendor): Cisco says DefenseClaw endpoint hooks fail closed by default, with Cisco Secure Client deployments defaulting to fail open when no policy is specified.Quote: "Only Cisco Secure Client deployments default hook delivery to fail-open when no policy is specified"Source: [github.com](https://github.com/cisco-ai-defense/defenseclaw), read 2026-11-16. - **cisco-10** (vendor): DefenseClaw documentation says its guardrail checks a prompt, a shell command or a tool call before the agent acts and answers allow, alert, ask or block.Quote: "The guardrail checks what your agent is about to do (a prompt, a shell command, a tool call) against rules, and answers allow, alert, ask you, or block."Source: [cisco-ai-defense.github.io](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/), read 2026-11-16. - **cisco-11** (vendor): DefenseClaw documentation says support on Linux, macOS and Windows is marked Supported only after a live test verified hooks and a block before the tool call ran.Quote: "means a live test on that OS verified the agent's hooks and a block before the tool call ran."Source: [cisco-ai-defense.github.io](https://cisco-ai-defense.github.io/defenseclaw/docs/support-matrix/), read 2026-11-16. - **cisco-12** (vendor): The DefenseClaw compatibility table lists hook contracts for Claude Code that inspect prompts, tool calls and tool results, with similar rows for Codex, Cursor, GitHub Copilot CLI and others.Quote: "Claude Code | `>=2.1.154, <2.1.219` | `claudecode-hooks-v1` / `v7` | 28 | prompt, tool_call, tool_result, event_content"Source: [cisco-ai-defense.github.io](https://cisco-ai-defense.github.io/defenseclaw/docs/connectors/compatibility/), read 2026-11-16. - **cisco-13** (vendor): On 2026-10-21, Cisco said DefenseClaw checks every prompt, tool call and tool result in an NVIDIA OpenShell sandbox using its own rules and Cisco AI Defense.Quote: "Every prompt, tool call and tool result is checked and observed by DefenseClaw’s rules and by Cisco AI Defense."Source: [blogs.cisco.com](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell), read 2026-11-16. - **cisco-14** (vendor): On 2026-10-21, Cisco said DefenseClaw decides each web destination for a sandboxed coding agent by host name without decrypting traffic, which covers outbound network calls.Quote: "DefenseClaw decides each web destination by host name, without decrypting the traffic."Source: [blogs.cisco.com](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell), read 2026-11-16. - **cisco-15** (vendor): On 2026-10-21, Cisco described the sandbox as supplying provider credentials only for authorized requests, so the agent sees placeholders instead of real credentials.Quote: "Provider credentials are supplied only for authorized requests; inside the sandbox, the agent sees placeholders."Source: [blogs.cisco.com](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell), read 2026-11-16. - **cisco-16** (vendor): On 2026-10-21, Cisco said enterprise administrators can set user- or group-based policy for which coding agents run on managed endpoints and which run only in a sandbox.Quote: "Enterprise administrators can configure user or group-based policy to determine which coding agents can run on their managed endpoints"Source: [blogs.cisco.com](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell), read 2026-11-16. - **cisco-17** (vendor): On 2026-10-21, Cisco said every block, approval and decision is recorded with the agent, sandbox and policy behind it and can be sent to Splunk or any OpenTelemetry backend.Quote: "Every block, approval and decision is recorded along with the agent, the sandbox and the policy behind it, and can be sent to Splunk or any OpenTelemetry backend."Source: [blogs.cisco.com](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell), read 2026-11-16. - **cisco-18** (vendor): DefenseClaw observability documentation lists OTLP, Splunk, Galileo, file and webhook destinations for logs, traces and metrics.Quote: "goes to the destinations you add: an OpenTelemetry (OTLP) collector, Splunk, Galileo, a file, or a webhook."Source: [cisco-ai-defense.github.io](https://cisco-ai-defense.github.io/defenseclaw/docs/observability/), read 2026-11-16. - **cisco-19** (vendor): DefenseClaw documentation says human approval is off by default, with a default minimum severity of HIGH when turned on.Quote: "Human approval is off by default. Its default minimum severity is HIGH."Source: [cisco-ai-defense.github.io](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/), read 2026-11-16. - **cisco-20** (vendor): DefenseClaw documentation states a limit: its command-text check cannot see what a child program writes after a command starts, so protected files need a sandbox or file permissions.Quote: "It cannot see what a child program writes after a command starts. For protected files, add a sandbox or file permissions."Source: [cisco-ai-defense.github.io](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/), read 2026-11-16. - **cisco-27** (vendor): The MCP Scanner runs as a stand-alone command-line tool or a REST API server, installed with uv.Quote: "Run scanner as a stand-alone CLI tool or REST API server"Source: [github.com](https://github.com/cisco-ai-defense/mcp-scanner), read 2026-11-16. - **cisco-29** (vendor): DefenseClaw documentation says its public website is the source of truth for install, configuration and operator workflows, and the code is Apache-2.0.Quote: "is the source of truth for installation, setup, configuration, commands, and operator workflows"Source: [github.com](https://github.com/cisco-ai-defense/defenseclaw), read 2026-11-16. - **cisco-31** (third-party): On 2026-03-23, SiliconANGLE reported DefenseClaw lets administrators block specific MCP servers, with block rules applied in two seconds without restarting agents.Quote: "A permission management feature enables administrators to block specific MCP servers."Source: [siliconangle.com](https://siliconangle.com/2026/03/23/cisco-debuts-new-ai-agent-security-features-open-source-defenseclaw-tool/), read 2026-11-16. - **cisco-33** (third-party): SiliconANGLE reported developers can install DefenseClaw in about five minutes and that it scans MCP tools and plug-ins an agent uses.Quote: "Cisco says developers can install DefenseClaw in about five minutes."Source: [siliconangle.com](https://siliconangle.com/2026/03/23/cisco-debuts-new-ai-agent-security-features-open-source-defenseclaw-tool/), read 2026-11-16. - **cisco-35** (vendor): Cisco says DefenseClaw streams every scan result, block decision, tool call and alert as structured events.The exact quote is on the source page.Source: [blogs.cisco.com](https://blogs.cisco.com/ai/defenseclaw-is-live), read 2026-11-16. - **cisco-38** (vendor): DefenseClaw documentation describes an enterprise edition that an administrator deploys with a package or an MDM and that users cannot turn off.Quote: "the managed standalone profile an administrator deploys with a package or an MDM. Users can't turn it off."Source: [cisco-ai-defense.github.io](https://cisco-ai-defense.github.io/defenseclaw/docs/support-matrix/), read 2026-11-16. - **cisco-39** (third-party): CRN reported at RSAC 2026 that Cisco plans to let customers register agents in Duo IAM mapped to human workers, with Identity Intelligence discovering agent identities.Quote: "register agents in Duo IAM, with the agents then mapped to accountable human workers"Source: [crn.com](https://www.crn.com/news/security/2026/cisco-unveils-zero-trust-for-ai-agents-5-things-to-know), read 2026-11-16. - **cisco-40** (third-party): SiliconANGLE reported on 2026-03-23 that Cisco's planned Duo IAM agent controls can limit an agent's tool access to specific time frames.Quote: "Agents’ tool access can be limited to specific time frames."Source: [siliconangle.com](https://siliconangle.com/2026/03/23/cisco-debuts-new-ai-agent-security-features-open-source-defenseclaw-tool/), read 2026-11-16. - **cisco-41** (vendor): Cisco says its planned MCP gateway applies authorization policies across tools and blocks out-of-bounds agent actions, but also says the features are in development.Quote: "the MCP gateway applies authorization policies consistently across tools and services."Source: [blogs.cisco.com](https://blogs.cisco.com/security/security-agentic-ai-how-cisco-brings-zero-trust-to-your-new-digital-workforce), read 2026-11-16. - **cisco-42** (vendor): Cisco says it discovers and registers AI agents, MCP servers and tools into a central inventory, with a note that many features are still in development.Quote: "Cisco discovers and registers AI agents, MCP servers, and associated tools"Source: [blogs.cisco.com](https://blogs.cisco.com/security/security-agentic-ai-how-cisco-brings-zero-trust-to-your-new-digital-workforce), read 2026-11-16. - **cisco-43** (vendor): Cisco says AI Cloud Visibility inventories AI models and connected data sources across a distributed environment.Quote: "Automatically inventory the AI models and connected data sources across your distributed environment to understand usage and gauge risk."Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-44** (vendor): The AI Defense Python SDK README says it can inspect AWS AgentCore and Strands event streams, releasing events only after server acknowledgement.Quote: "Inspect AgentCore, Strands, and adapter-based provider streams, releasing events only after server acknowledgement."Source: [github.com](https://github.com/cisco-ai-defense/ai-defense-python-sdk), read 2026-11-16. - **cisco-45** (observed): The GitHub org lists a Cisco AI Defense plugin for Google ADK that does callback-based LLM and MCP inspection, plus LangChain and LiteLLM integrations.The exact quote is on the source page.Source: [api.github.com](https://api.github.com/orgs/cisco-ai-defense/repos), read 2026-11-16. - **cisco-46** (none-found): None found on 2026-11-16: named coverage of Microsoft Copilot Studio, Salesforce Agentforce or ServiceNow agents on the AI Defense product page, Kong plugin, SDK or DefenseClaw docs read.Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-47** (none-found): None found on 2026-11-16: coverage of browser agents, AI browsers or AI browser extensions on the AI Defense product page or DefenseClaw connector and support pages.Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-48** (vendor): Cisco AI Access, part of Secure Access SSE, inspects web traffic to identify shadow AI use; it is network-based rather than a browser-agent control.Quote: "AI Access inspects web traffic to identify uses of shadow AI across your organization."Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/secure-access/ai-access/index.html), read 2026-11-16. - **cisco-53** (vendor): On 2026-10-28, Cisco said its Project Quarry detection language is in production in AI Defense and maps findings to alert, block, redact, quarantine or log.Quote: "Call Quarry from the application path and map findings to alert, block, redact, quarantine, or log."Source: [blogs.cisco.com](https://blogs.cisco.com/ai/project-quarry-agent-security-detections), read 2026-11-16. - **cisco-54** (vendor): On 2026-11-09, Cisco published a blog titled 'Cisco AI Defense to Add Anthropic's Claude Mythos for AI Red Teaming'.Quote: "Cisco AI Defense to Add Anthropic’s Claude Mythos for AI Red Teaming"Source: [blogs.cisco.com](https://blogs.cisco.com/ai/project-quarry-agent-security-detections), read 2026-11-16. - **cisco-55** (third-party): On 2026-10-26, SiliconANGLE reported Cisco launched Privileged Infrastructure Access, applying task-scoped rules to AI agents as to people, after an August Teleport partnership and stake.Quote: "AI agents and automation scripts work under the same task-scoped rules as people"Source: [siliconangle.com](https://siliconangle.com/2026/10/26/ciscos-privileged-infrastructure-access-ties-admin-and-ai-agent-privileges-to-the-task/), read 2026-11-16. - **cisco-56** (none-found): None found on 2026-11-16: a published price for Cisco AI Defense on its product page; the free Explorer Edition is announced but no paid list price was found.Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-57** (third-party): CRN reported that at RSAC 2026 Cisco introduced a free Explorer Edition of AI Defense for red teaming agents.Quote: "Cisco also debuted updates to its AI Defense offering with the introduction of a free “Explorer Edition.”"Source: [crn.com](https://www.crn.com/news/security/2026/cisco-unveils-zero-trust-for-ai-agents-5-things-to-know), read 2026-11-16. - **cisco-58** (vendor): The AI Defense product page quotes Kent Noyes of World Wide Technology, a named customer voice.Quote: "The adoption of AI exposes companies to new risks that traditional cybersecurity solutions don't address."Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-59** (none-found): None found on 2026-11-16: independent analyst coverage (Gartner, Forrester, IDC) or a funding round for AI Defense on the pages read; Cisco is a public company.Source: [cisco.com](https://www.cisco.com/site/us/en/products/security/ai-defense/index.html), read 2026-11-16. - **cisco-60** (third-party): SiliconANGLE covered the 2026-03-23 DefenseClaw release at RSAC 2026 and the Explorer Edition.Quote: "The networking giant also released an open-source tool, DefenseClaw, that can scan AI agents for vulnerabilities."Source: [siliconangle.com](https://siliconangle.com/2026/03/23/cisco-debuts-new-ai-agent-security-features-open-source-defenseclaw-tool/), read 2026-11-16. ## News mentioning Cisco - [Cisco DefenseClaw runs coding agents in a sandbox](/news/cisco-defenseclaw-nvidia-openshell-coding-agents/) ## Related pages - [All vendor profiles](/vendors/) - [Check Point profile](/vendors/check-point/) - [CrowdStrike profile](/vendors/crowdstrike/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # CrowdStrike AI agent security URL: https://aiagentsecuritysolutions.com/vendors/crowdstrike/ Updated: 2026-11-16 CrowdStrike spans the endpoint, identity and gateway layers. On 3 Nov 2026 it said agent-level hooks cover Claude Code, OpenAI Codex, GitHub Copilot and OpenCode and block before actions run, and that an AI Kill Switch in Falcon Guardian directs connected AI gateways. Its Agentic Identity Provider, announced 2 Sep 2026, registers agents and brokers short-lived, scoped access that can be revoked, and a third-party report confirms it. Falcon AI Gateway is included with Guardian at no added cost. Cloud agent discovery is pre-beta or coming soon, the wider kill switch is not yet available, and the agent product has no published price. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers the most layers of any single stack: endpoint sensor and hooks, a generally available gateway with kill switch, and an agent identity provider (facts 10, 24, 31). **Weakness:** Cloud agent coverage is pre-beta or coming soon, and the wider kill switch is not yet available (facts 25, 26, 27). ## Scores and evidence CrowdStrike is 6th in the overall ranking at 60.3 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **CrowdStrike: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 7 of 10 | 6 | Sensor visibility on three OSes and hooks for named coding agents that block before actions run are described; vendor claims only. | [crowdstrike-02](#fact-crowdstrike-02), [crowdstrike-09](#fact-crowdstrike-09), [crowdstrike-10](#fact-crowdstrike-10), [crowdstrike-29](#fact-crowdstrike-29) | | Coverage: browser layer | 5 of 10 | 6 | An enterprise browser and extension controls inventory extensions and enforce paste and upload policy; browser agent control none found. | [crowdstrike-35](#fact-crowdstrike-35), [crowdstrike-36](#fact-crowdstrike-36), [crowdstrike-37](#fact-crowdstrike-37) | | Coverage: cloud workspace layer | 1 of 10 | 6 | Cloud and Bedrock AgentCore discovery are pre-beta or coming soon only. | [crowdstrike-26](#fact-crowdstrike-26), [crowdstrike-27](#fact-crowdstrike-27), [crowdstrike-28](#fact-crowdstrike-28) | | Coverage: identity layer | 8 of 10 | 6 | An agentic IdP registers agents, brokers short-lived scoped access and revokes it, confirmed by a third-party report. | [crowdstrike-31](#fact-crowdstrike-31), [crowdstrike-32](#fact-crowdstrike-32), [crowdstrike-33](#fact-crowdstrike-33), [crowdstrike-34](#fact-crowdstrike-34) | | Coverage: gateway layer | 7 of 10 | 6 | A central AI gateway inspects prompts and responses and a kill switch blocks model access; it acts only on routed traffic. | [crowdstrike-18](#fact-crowdstrike-18), [crowdstrike-19](#fact-crowdstrike-19), [crowdstrike-21](#fact-crowdstrike-21), [crowdstrike-24](#fact-crowdstrike-24) | | Stops the action | 7 of 10 | 20 | The vendor blocks malicious agent activity, unauthorized agent types and malicious packages before execution; vendor claims only. | [crowdstrike-10](#fact-crowdstrike-10), [crowdstrike-11](#fact-crowdstrike-11), [crowdstrike-12](#fact-crowdstrike-12), [crowdstrike-16](#fact-crowdstrike-16) | | Visibility | 7 of 10 | 15 | Discovery of agents, MCP servers and skills plus session reconstruction linking tool calls to identity is described; vendor claims only. | [crowdstrike-01](#fact-crowdstrike-01), [crowdstrike-09](#fact-crowdstrike-09), [crowdstrike-40](#fact-crowdstrike-40), [crowdstrike-41](#fact-crowdstrike-41) | | Deployment | 6 of 10 | 10 | Sensor and device-management rollout on three OSes is described; a time to deploy none found. | [crowdstrike-29](#fact-crowdstrike-29), [crowdstrike-30](#fact-crowdstrike-30), [crowdstrike-55](#fact-crowdstrike-55), [crowdstrike-60](#fact-crowdstrike-60) | | Policy and governance | 6 of 10 | 10 | Agent-type policy, SIEM export and managed response are described; ticketing and third-party SIEM integrations none found. | [crowdstrike-12](#fact-crowdstrike-12), [crowdstrike-38](#fact-crowdstrike-38), [crowdstrike-39](#fact-crowdstrike-39), [crowdstrike-42](#fact-crowdstrike-42) | | Transparency | 3 of 10 | 10 | Research and endpoint bundle prices are public; a price for the agent product and public documentation none found. | [crowdstrike-45](#fact-crowdstrike-45), [crowdstrike-46](#fact-crowdstrike-46), [crowdstrike-47](#fact-crowdstrike-47), [crowdstrike-48](#fact-crowdstrike-48) | | Proof | 8 of 10 | 5 | Third parties report acquisitions in the space, plus a customer story and an IDC MarketScape leader listing. | [crowdstrike-05](#fact-crowdstrike-05), [crowdstrike-43](#fact-crowdstrike-43), [crowdstrike-44](#fact-crowdstrike-44), [crowdstrike-63](#fact-crowdstrike-63) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 6th | 60.3 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | joint 5th | 64.5 | | [Ranked for the browser layer](/best-for/layer-browser/) | 7th | 58.5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 14th | 46.5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 1st | 67.5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 4th | 64.5 | ## Product, deployment, price and owner - Product assessed: Falcon Guardian (AIDR) with Falcon AI Gateway and AI Kill Switch, plus Agentic Identity Provider and Falcon Seraphic Enterprise Browser for identity and browser layers - Deployment: Falcon sensor on endpoints with agent-level hooks, a routed Falcon AI Gateway for apps and workloads beyond the sensor, and device-management rollout for the Cowork integration - Time to deploy: none found on 2026-11-16 - Price: none found on 2026-11-16 for Falcon Guardian; endpoint bundles are priced (Falcon Go $7.99 per device monthly) and Falcon AI Gateway is included with Guardian at no additional cost - Owner: CrowdStrike Holdings, Inc. (public company; CrowdStrike sells the products itself) - Website: crowdstrike.com ## Facts and sources - **crowdstrike-01** (vendor): CrowdStrike says Falcon Guardian discovers shadow AI, agents, applications, models, MCP servers, skills and tools on endpoints.Quote: "Discover shadow AI, agents, applications, models, MCP servers, skills, and tools."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/platform/falcon-guardian-aidr/), read 2026-11-16. - **crowdstrike-02** (vendor): On 2026-09-01 CrowdStrike introduced Falcon Guardian, the successor to Falcon AI Detection and Response (AIDR), and says it discovers AI agents on supported Windows, macOS and Linux endpoints.Quote: "Falcon Guardian continuously discovers known and previously unknown AI agents across supported Windows, macOS, and Linux endpoints."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-03** (vendor): CrowdStrike names Falcon Guardian as the evolution of Falcon AI Detection and Response (AIDR) and its flagship AI security product.Quote: "the evolution of Falcon AI Detection and Response (AIDR) and CrowdStrike’s flagship solution for the AIDR category"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-04** (vendor): CrowdStrike lists George Kurtz as CEO and Founder and Michael Sentonas as President on its executive team page.Quote: "George Kurtz CEO & Founder Michael Sentonas President"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/about-us/executive-team/), read 2026-11-16. - **crowdstrike-05** (third-party): SiliconANGLE reported on 2026-11-03 that CrowdStrike Holdings plans to buy Fiddler AI and Dreadnode, two AI agent security startups, announced with a Falcon Guardian update.Quote: "CrowdStrike Holdings Inc. today disclosed plans to buy Fiddler AI Inc. and Dreadnode Inc., two startups that specialize in securing artificial intelligence agents."Source: [siliconangle.com](https://siliconangle.com/2026/11/03/crowdstrike-to-acquire-cybersecurity-startups-fiddler-ai-dreadnode-for-reported-495m/), read 2026-11-16. - **crowdstrike-06** (vendor): On 2026-11-03 CrowdStrike announced its intent to acquire Fiddler, to become Falcon Observe for AI agent observability, generally available in Q4 FY27 (a future product, not scored as available).Quote: "Fiddler will become part of the Falcon platform as Falcon Observe, generally available in Q4 FY27."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-to-acquire-fiddler/), read 2026-11-16. - **crowdstrike-07** (third-party): SiliconANGLE says Dreadnode, being acquired by CrowdStrike and planned for integration into Falcon Guardian, looks for vulnerabilities in the MCP servers an agent uses (planned, not yet part of Guardian).Quote: "Dreadnode also looks for vulnerabilities in the MCP servers that an AI agent uses to carry out its work."Source: [siliconangle.com](https://siliconangle.com/2026/11/03/crowdstrike-to-acquire-cybersecurity-startups-fiddler-ai-dreadnode-for-reported-495m/), read 2026-11-16. - **crowdstrike-09** (vendor): On 2026-11-03 CrowdStrike said the Falcon sensor gives OS-level visibility into agent processes, commands and resource access, available now.Quote: "The Falcon sensor provides deep visibility into agents’ activity, including the processes they launch, commands they run, and resources they access."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/), read 2026-11-16. - **crowdstrike-10** (vendor): On 2026-11-03 CrowdStrike said agent-level hooks for Claude Code, OpenAI Codex, GitHub Copilot, OpenCode and Cursor inspect prompts and block malicious activity before harmful actions run (hooks act on the agent, not specifically on MCP tool calls).Quote: "Hooks for Claude Code, OpenAI Codex, GitHub Copilot, OpenCode, and Cursor enable organizations to inspect prompts and block malicious activity at the source, before harmful actions are executed."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/), read 2026-11-16. - **crowdstrike-11** (vendor): On 2026-09-01 CrowdStrike said Falcon Guardian can drive automatic containment by blocking malicious agent behaviors and compromised assets at runtime.Quote: "drive automatic containment by blocking malicious agent behaviors and compromised assets at runtime across agents"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-12** (vendor): On 2026-09-01 CrowdStrike said Falcon Guardian lets organizations define which supported agent types may run on managed endpoints and blocks unauthorized types.Quote: "helping security teams sanction approved agents and prevent unauthorized agent types from running."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-13** (vendor): CrowdStrike says Falcon Guardian protects supported AI interactions against prompt injection and sensitive data exposure.Quote: "Existing Falcon Guardian controls continue to protect supported AI interactions against threats such as prompt injection and sensitive data exposure."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-15** (vendor): On 2026-11-03 CrowdStrike said Falcon Guardian discovers malicious AI skills and MCP servers to identify threats before they enter the AI environment.Quote: "Falcon Guardian expands AI supply chain protection by discovering malicious AI skills and MCP servers to help security teams identify threats before they enter the AI environment."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/), read 2026-11-16. - **crowdstrike-16** (vendor): On 2026-09-02 CrowdStrike said the Falcon sensor blocks malicious open-source packages as they reach the endpoint, before embedded code executes (a sensor capability, separate from Guardian).Quote: "It detects and blocks malicious open-source packages as they reach the endpoint, before any embedded code can execute"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-extends-endpoint-security-to-stop-supply-chain-attacks/), read 2026-11-16. - **crowdstrike-18** (vendor): On 2026-11-13 CrowdStrike said Falcon AI Gateway routes AI traffic through a central gateway that inspects prompts and responses and enforces policy on managed and unmanaged devices.Quote: "It routes AI traffic through a centralized gateway to inspect prompts and responses, monitor usage, and enforce policy across managed and unmanaged devices."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/), read 2026-11-16. - **crowdstrike-19** (vendor): CrowdStrike states the gateway acts only on traffic routed through it and cannot stop local processes.Quote: "Falcon AI Gateway acts on traffic routed through it and can't stop local processes."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/), read 2026-11-16. - **crowdstrike-20** (vendor): CrowdStrike states the gateway does not discover or redirect AI traffic automatically; teams choose which workloads to route through it.Quote: "This doesn't discover or redirect AI traffic automatically; it gives teams a deliberate control point for the workloads they choose to route through it."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/), read 2026-11-16. - **crowdstrike-21** (vendor): CrowdStrike says developers connect an existing SDK or coding tool to the gateway by changing the base URL and using a gateway key.Quote: "Developers connect an existing SDK or coding tool by changing the base URL to point at the gateway and using a gateway key."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/), read 2026-11-16. - **crowdstrike-23** (vendor): On 2026-11-03 CrowdStrike said Falcon AI Gateway is included with Falcon Guardian at no additional cost.Quote: "Included with Falcon Guardian at no additional cost, it extends AI security to applications and workloads beyond direct endpoint sensor coverage, including unmanaged environments."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/), read 2026-11-16. - **crowdstrike-24** (vendor): On 2026-11-03 CrowdStrike said AI Kill Switch in Falcon Guardian directs connected AI gateways (Falcon or third-party) to block access to the models that power agents.Quote: "Security teams activate AI Kill Switch in Falcon Guardian, which directs connected AI gateways to block access to the AI models that power agents' decisions and actions."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/), read 2026-11-16. - **crowdstrike-25** (vendor): On 2026-11-03 CrowdStrike described as coming soon (not available) a wider AI Kill Switch spanning endpoint, cloud, identity, SaaS and browser enforcement.Quote: "Coming soon, CrowdStrike will extend AI Kill Switch beyond gateway enforcement to deliver coordinated agent containment across the Falcon platform."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/), read 2026-11-16. - **crowdstrike-26** (vendor): On 2026-11-03 CrowdStrike described pre-beta discovery for provider-managed agents, starting with Amazon Bedrock AgentCore.Quote: "Falcon Guardian will expand agent discovery and AI-layer visibility, beginning with Amazon Bedrock AgentCore."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/), read 2026-11-16. - **crowdstrike-27** (vendor): On 2026-11-03 CrowdStrike described pre-beta coverage of cloud workload and Kubernetes agents, including LangChain and LangGraph apps.Quote: "In cloud workloads and Kubernetes, Falcon Guardian will extend these capabilities to enterprise-developed AI agents and applications, including those built with LangChain and LangGraph."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/), read 2026-11-16. - **crowdstrike-28** (vendor): CrowdStrike lists discovery across cloud workloads, Linux, Kubernetes and managed AI platforms as coming soon on the Guardian page.Quote: "Coming soon: discover agents across cloud workloads, Linux, Kubernetes, and managed AI platforms, extending visibility beyond the endpoint."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/platform/falcon-guardian-aidr/), read 2026-11-16. - **crowdstrike-29** (vendor): On 2026-11-03 CrowdStrike said its Claude Cowork integration is in beta on macOS, with Windows expected later in November.Quote: "The integration is now available in beta on macOS, with Windows support expected later in November."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-secures-ai-agents-inside-claude-cowork/), read 2026-11-16. - **crowdstrike-30** (vendor): On 2026-11-03 CrowdStrike said admins turn on the Cowork integration through existing device management, as they deploy the Falcon sensor.Quote: "Admins will turn the integration on through their existing device management, the same way they deploy the Falcon sensor today"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-secures-ai-agents-inside-claude-cowork/), read 2026-11-16. - **crowdstrike-31** (vendor): On 2026-09-02 CrowdStrike said its Agentic Identity Provider brokers short-lived, scoped access for agents instead of standing credentials.Quote: "Instead of assigning agents standing credentials of their own, the Agentic Identity Provider securely brokers short-lived, tightly scoped access on their behalf"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/), read 2026-11-16. - **crowdstrike-32** (vendor): On 2026-09-02 CrowdStrike said the Agentic IdP registers agents in one directory and gives each a cryptographically verifiable identity.Quote: "The Agentic Identity Provider automatically registers agents in a single directory and gives each one a cryptographically verifiable identity."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/), read 2026-11-16. - **crowdstrike-33** (vendor): CrowdStrike says Continuous Identity grants agent access when justified and revokes it when conditions change.Quote: "This capability grants access when it is justified and revokes it the moment conditions change."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/), read 2026-11-16. - **crowdstrike-34** (third-party): SiliconANGLE reported on 2026-09-02 that agent registration for the Agentic IdP runs through Falcon Guardian.Quote: "Registration runs through Falcon Guardian, the agent discovery and enforcement product CrowdStrike launched"Source: [siliconangle.com](https://siliconangle.com/2026/09/02/crowdstrike-gives-ai-agents-an-identity-provider-parallel-soc-investigations-and-package-blocking/), read 2026-11-16. - **crowdstrike-35** (vendor): On 2026-11-04 CrowdStrike said Falcon Seraphic Enterprise Browser lets teams inventory browser extensions, rate their risk and allow or block them.Quote: "Security teams can inventory browser extensions, evaluate their risk, and enforce policies that allow or block risky extensions across supported browsers."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-brings-enterprise-browser-security-to-falcon-platform/), read 2026-11-16. - **crowdstrike-36** (vendor): On 2026-11-04 CrowdStrike said Seraphic can enforce copy, paste, upload and download policies to keep sensitive data from unsanctioned web apps and generative AI tools (a separate product from Guardian).Quote: "Teams can enforce policies on browser actions such as copy, paste, upload, and download; this helps prevent sensitive data from reaching unsanctioned web applications and generative AI tools."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-brings-enterprise-browser-security-to-falcon-platform/), read 2026-11-16. - **crowdstrike-37** (vendor): On 2026-11-04 CrowdStrike said Seraphic protection works in existing browsers with no dedicated browser or traffic rerouting.Quote: "Protection works in the browsers teams already use; there is no need to deploy a dedicated browser or reroute traffic."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-brings-enterprise-browser-security-to-falcon-platform/), read 2026-11-16. - **crowdstrike-38** (vendor): On 2026-09-01 CrowdStrike said Guardian exports agent telemetry into Falcon Next-Gen SIEM as first-party data.Quote: "Falcon Guardian natively exports agent telemetry into CrowdStrike Falcon® Next-Gen SIEM as first-party data, pre-mapped to its schema for immediate correlation, detection, and automation"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-39** (vendor): On 2026-11-03 CrowdStrike listed Guardian additions: Falcon Data Replicator export of prompt activity, Flight Control support, and token consumption dashboards.Quote: "Falcon Data Replicator integration for prompt activity, Falcon Flight Control support, new executive summary and token consumption dashboards, and expanded Windows AI application coverage."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/), read 2026-11-16. - **crowdstrike-40** (vendor): On 2026-09-01 CrowdStrike said Guardian reconstructs agent sessions and downstream execution into a causal investigation.Quote: "Falcon Guardian reconstructs agent sessions and downstream execution into a unified causal investigation, allowing analysts to trace suspicious activity across affected agents and systems."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-41** (vendor): CrowdStrike says Guardian links prompts, skill use, tool calls, MCP servers and identity to downstream execution for supported agents.Quote: "connect prompts, skill use, tool calls, MCP servers, and identity with downstream system execution for supported agents."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/), read 2026-11-16. - **crowdstrike-42** (vendor): On 2026-11-03 CrowdStrike announced Falcon Complete for Guardian is generally available, adding 24/7 managed detection and response for AI security.Quote: "the general availability of Falcon Complete for Guardian, extending managed detection and response (MDR) services to AI security."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/), read 2026-11-16. - **crowdstrike-43** (vendor): CrowdStrike publishes a customer story saying Cogentrix gained day-one AI visibility with Falcon Guardian (customer claim, no time figure).Quote: "See how Cogentrix gains day-one AI visibility and governs employee and agentic AI with Falcon Guardian."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/platform/falcon-guardian-aidr/), read 2026-11-16. - **crowdstrike-44** (vendor): CrowdStrike lists on its blog that it was named a Leader in the 2026 IDC MarketScape for Worldwide Augmented Identity Security Solutions (dated 2026-11-12; identity security, not AI security specifically).Quote: "CrowdStrike Named a Leader in the 2026 IDC MarketScape for Worldwide Augmented Identity Security Solutions"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/), read 2026-11-16. - **crowdstrike-45** (vendor): CrowdStrike publishes list prices for endpoint bundles, for example Falcon Go at $7.99 per device billed monthly; Guardian is not on that page.Quote: "Falcon Go $7.99 per device, billed monthly"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/pricing/), read 2026-11-16. - **crowdstrike-46** (none-found): None found on 2026-11-16: a published price for Falcon Guardian or Falcon AIDR on the pricing page (only endpoint bundles are priced; other products say contact us).Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/pricing/), read 2026-11-16. - **crowdstrike-47** (none-found): None found on 2026-11-16: public product documentation for Falcon Guardian; the documentation address returned a Falcon login page.Source: [falcon.crowdstrike.com](https://falcon.crowdstrike.com/documentation/), read 2026-11-16. - **crowdstrike-48** (vendor): On 2026-01-09 CrowdStrike published research explaining AI tool poisoning through MCP tool descriptions and how to defend against it.Quote: "In this blog, we explain how AI tool poisoning works, the different forms it can take, and how organizations can strengthen their defenses against this type of attack."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/ai-tool-poisoning/), read 2026-11-16. - **crowdstrike-49** (vendor): CrowdStrike links an interactive Taxonomy of Prompt Injection Methods as published research from its AI agent security page.Quote: "Taxonomy of Prompt Injection Methods"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/solutions/ai-agent-security/), read 2026-11-16. - **crowdstrike-53** (vendor): CrowdStrike sells a Shadow AI Visibility Service (expert-led, launched 2026-04-21) that discovers AI tools, agents, copilots and extensions across endpoint, cloud and SaaS.Quote: "Discover AI tools, agents, copilots, extensions, and model-connected services operating across endpoint, cloud, and SaaS"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/services/ai-security-services/shadow-ai-visibility/), read 2026-11-16. - **crowdstrike-54** (vendor): On 2026-11-04 CrowdStrike described a Q4 roadmap feature in Next-Gen SIEM that baselines each agent and its owner and flags deviations (not yet available).Quote: "learns the normal behavior pattern for every AI agent and the person who runs it, the files it touches, the tools it calls"Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/blog/crowdstrike-detects-and-stops-agent-insider-threats/), read 2026-11-16. - **crowdstrike-55** (vendor): CrowdStrike says endpoint network inspection adds traffic-level controls while the Falcon sensor extends control to runtime on the endpoint.Quote: "Endpoint Network Inspection adds traffic-level controls, while the Falcon sensor extends control to runtime on the endpoint."Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/platform/falcon-guardian-aidr/), read 2026-11-16. - **crowdstrike-60** (none-found): None found on 2026-11-16: a published time to deploy for Falcon Guardian on the Guardian page.Source: [crowdstrike.com](https://www.crowdstrike.com/en-us/platform/falcon-guardian-aidr/), read 2026-11-16. - **crowdstrike-63** (third-party): TechRadar reported that CrowdStrike confirmed it would acquire identity security startup SGNL for about $740 million, to support continuous access control for human, non-human and AI identities (article undated in extracted text).Quote: "CrowdStrike has confirmed it will be acquiring identity security startup SGNL for around $740 million in a mostly-cash deal, subject to regulatory approval."Source: [techradar.com](https://www.techradar.com/pro/crowdstrike-acquires-continuous-identity-firm-sgnl-for-usd740m-in-push-for-next-gen-security), read 2026-11-16. ## News mentioning CrowdStrike - [CrowdStrike Falcon AI Gateway reaches general availability](/news/crowdstrike-falcon-ai-gateway-generally-available/) - [CrowdStrike adds Seraphic browser security to Falcon](/news/crowdstrike-seraphic-enterprise-browser-falcon/) - [CrowdStrike launches an identity provider for AI agents](/news/crowdstrike-agentic-identity-provider/) ## Related pages - [All vendor profiles](/vendors/) - [Cisco profile](/vendors/cisco/) - [Island profile](/vendors/island/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Island AI agent security URL: https://aiagentsecuritysolutions.com/vendors/island/ Updated: 2026-11-16 Island builds agent security around its enterprise browser. Its pages describe a headless browser for agents with credential isolation, a live device inventory of MCP servers, clients and CLI agents with risk-based restriction, and an MCP Gateway that issues just-in-time credentials scoped to one task and limits an agent per tool, action and actor. Non-human identity inventory ties to Okta and Entra revocation steps, and one policy engine covers the surfaces with audit logs. Third parties report large funding and customers. Cloud platform agents are covered only through Claude surfaces via the Compliance API, and a price and time to deploy: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers endpoint, browser, identity and gateway layers under one policy engine with inline blocking at several control points (island-13, island-15, island-06). **Weakness:** No Copilot Studio, Agentforce, ServiceNow or Bedrock agent coverage, no time to deploy and no price found on 2026-11-16 (island-10, island-23). ## Scores and evidence Island is 7th in the overall ranking at 60.1 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Island: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 7 of 10 | 6 | A live device inventory of MCP servers, clients and CLI agents with risk-based restriction is described; vendor claims only. | [island-04](#fact-island-04), [island-05](#fact-island-05), [island-06](#fact-island-06) | | Coverage: browser layer | 7 of 10 | 6 | A headless browser for agents, AI browser support and credential isolation are described; vendor claims only. | [island-07](#fact-island-07), [island-08](#fact-island-08), [island-09](#fact-island-09) | | Coverage: cloud workspace layer | 3 of 10 | 6 | Only Claude surfaces via the Compliance API are covered; Copilot Studio, Agentforce, ServiceNow and Bedrock none found. | [island-10](#fact-island-10), [island-11](#fact-island-11) | | Coverage: identity layer | 7 of 10 | 6 | Non-human identity inventory, just-in-time scoped credentials and Okta and Entra revocation steps are described; vendor claims only. | [island-12](#fact-island-12), [island-13](#fact-island-13), [island-14](#fact-island-14) | | Coverage: gateway layer | 7 of 10 | 6 | An MCP gateway limits agents per tool, action and actor; vendor claims only. | [island-13](#fact-island-13), [island-15](#fact-island-15), [island-16](#fact-island-16) | | Stops the action | 7 of 10 | 20 | Inline points stop actions, restrict access, apply read-only mode and block risky MCP servers; vendor claims only. | [island-06](#fact-island-06), [island-09](#fact-island-09), [island-16](#fact-island-16), [island-17](#fact-island-17) | | Visibility | 7 of 10 | 15 | Device inventory, full tool-call reading and audit logs are described; vendor claims only. | [island-04](#fact-island-04), [island-18](#fact-island-18), [island-19](#fact-island-19), [island-25](#fact-island-25) | | Deployment | 5 of 10 | 10 | Server-to-server, extension and EDR-based collection are described; a time to deploy and OS list none found. | [island-20](#fact-island-20), [island-21](#fact-island-21), [island-22](#fact-island-22) | | Policy and governance | 5 of 10 | 10 | One policy across surfaces, audit logs and Okta, Entra and EDR integrations exist; SIEM, SOAR and ticketing export none found. | [island-14](#fact-island-14), [island-16](#fact-island-16), [island-19](#fact-island-19), [island-30](#fact-island-30) | | Transparency | 3 of 10 | 10 | Research is public; a price and public documentation none found. | [island-23](#fact-island-23), [island-24](#fact-island-24) | | Proof | 8 of 10 | 5 | Third parties report large funding and customers, and the vendor cites a Gartner Peer Insights recognition. | [island-03](#fact-island-03), [island-26](#fact-island-26), [island-27](#fact-island-27), [island-28](#fact-island-28) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 7th | 60.1 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | joint 8th | 62.5 | | [Ranked for the browser layer](/best-for/layer-browser/) | joint 3rd | 62.5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 12th | 50.5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 5th | 62.5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 6th | 62.5 | ## Product, deployment, price and owner - Product assessed: Island Enterprise Agentic Control Plane: Agentic Endpoint Posture, Agentic Identity, AI Runtime Protection, Island MCP Gateway, Enterprise Browser for Agents, Island Extension and Island Desktop - Deployment: Island Desktop, browser and extension plus MCP Gateway, network and server-to-server API integrations (Claude Compliance API, Inference Hooks) - Time to deploy: none found on 2026-11-16 - Price: none found on 2026-11-16 (request a quote) - Owner: Island Technology Ltd. (standalone, privately held; Series F announced 2026-09-24) - Website: [island.io](https://www.island.io) ## Facts and sources - **island-01** (vendor): Island's home page title names the platform the Frontier Control Plane and says it is built for the agentic enterprise; products are the Enterprise Browser, Enterprise AI and Enterprise Network.Quote: "The control plane for your agentic enterprise"Source: [island.io](https://www.island.io), read 2026-11-16. - **island-02** (vendor): Island's about page lists Mike Fey as co-founder and CEO and Dan Amiga as co-founder and CTO.Quote: "Mike Fey Co-Founder & CEO Dan Amiga Co-Founder & CTO"Source: [island.io](https://www.island.io/about), read 2026-11-16. - **island-03** (vendor): Island says on 2026-09-24 it announced a $400 million Series F at a $6.4 billion valuation led by Evolution Equity Partners, and calls itself the agentic control plane for enterprises.Quote: "Evolution Equity Partners led the round, with participation from existing and new investors."Source: [island.io](https://www.island.io/press/island-announces-400-million-series-f-bringing-valuation-to-6-4-billion), read 2026-11-16. - **island-04** (vendor): Island says Agentic Endpoint Posture keeps a live inventory of MCP servers, AI clients, skills, hooks, code packages and IDE and browser extensions on every device.Quote: "A live inventory of every AI asset on every device. MCP servers, both configured and actively listening."Source: [island.io](https://www.island.io/product/ai/agentic-endpoint-posture), read 2026-11-16. - **island-05** (vendor): Island names the desktop clients, coding assistants and CLI agents its endpoint inventory recognizes.Quote: "including Claude Desktop, Claude Code, Claude Cowork, the ChatGPT desktop app, Codex CLI, the Microsoft 365 Copilot app, Cursor, Gemini CLI, and GitHub Copilot"Source: [island.io](https://www.island.io/product/ai/agentic-endpoint-posture), read 2026-11-16. - **island-06** (vendor): Island says endpoint posture enforces policy at once, restricting skills and MCP servers above a risk threshold and allow-listing sanctioned AI clients.Quote: "Restrict access to skills, MCP servers, and other assets above a defined risk threshold"Source: [island.io](https://www.island.io/product/ai/agentic-endpoint-posture), read 2026-11-16. - **island-07** (vendor): Island says its Enterprise Browser for Agents is a headless Island browser exposed to any MCP-capable agent, under per-agent, per-application policy and audit.Quote: "It is a headless version of the Island Enterprise Browser with its tools opened to AI agents."Source: [island.io](https://www.island.io/product/ai/enterprise-browser-for-agents), read 2026-11-16. - **island-08** (vendor): Island says its extension supports AI browsers such as ChatGPT Atlas and Perplexity Comet with the same feature support as in Chrome and Edge.Quote: "Island Extension supports AI browsers including Atlas and Comet, with the same feature support it has in Chrome and Edge"Source: [island.io](https://www.island.io/product/ai/enterprise-browser-for-agents), read 2026-11-16. - **island-09** (vendor): Island says it applies read-only mode per application and isolates credentials, so a vault fills login forms and agents never hold passwords.Quote: "The vault fills the form, so nobody hands an agent a password"Source: [island.io](https://www.island.io/product/ai/enterprise-browser-for-agents), read 2026-11-16. - **island-10** (none-found): None found on 2026-11-16: looked for Copilot Studio, Agentforce, ServiceNow and Bedrock agent coverage on the Island AI overview, agentic identity and runtime protection pages; only Claude surfaces (via the Compliance API) and Salesforce, AWS and GitHub identity inventory were named.Source: [island.io](https://www.island.io/product/ai), read 2026-11-16. - **island-11** (vendor): Island says its Claude Compliance API integration now reads Claude Code, Cowork and Claude for Microsoft 365 sessions as full transcripts, on 2026-09-29.Quote: "Anthropic has now extended the Compliance API to all of it: Claude Code, Claude Cowork, the Claude for Microsoft 365 add-ins"Source: [island.io](https://www.island.io/blog/claude-code-cowork-and-claude-for-microsoft-365-now-report-into-island), read 2026-11-16. - **island-12** (vendor): Island says Agentic Identity inventories non-human identities (service accounts, API keys, tokens, OAuth grants) from cloud, repositories, SaaS and AI tools.Quote: "Island inventories every non-human identity and governs their access."Source: [island.io](https://www.island.io/product/ai/agentic-identity), read 2026-11-16. - **island-13** (vendor): Island says its MCP Gateway issues just-in-time credentials scoped to one task so the agent never sees the underlying key.Quote: "Just-in-time access scoped to a single task, expiring when the session ends"Source: [island.io](https://www.island.io/product/ai/agentic-identity), read 2026-11-16. - **island-14** (vendor): Island says it integrates with Okta and Microsoft Entra to surface revocation and rotation steps for risky agent credentials.Quote: "Island integrates with identity providers like Okta and Microsoft Entra, and with application systems like Salesforce and GitHub"Source: [island.io](https://www.island.io/product/ai/agentic-identity), read 2026-11-16. - **island-15** (vendor): Island says its MCP Gateway sits between every agent and the tools it calls and can limit an agent per tool, per action and per actor.Quote: "The Island MCP Gateway sits between every agent and the tools it calls."Source: [island.io](https://www.island.io/product/ai/agentic-identity), read 2026-11-16. - **island-16** (vendor): Island says one policy applies across browser, extension, desktop, agent hooks, MCP Gateway, network and LLM gateways, and that the inline points can stop an action before it happens.Quote: "so they can stop an action before it happens. Provider APIs, OpenTelemetry, and local collection report after the event"Source: [island.io](https://www.island.io/product/ai/ai-runtime-protection), read 2026-11-16. - **island-17** (vendor): Island says AI Runtime Protection responds by policy in the moment: allow and log, prompt the user, redirect the agent or restrict access.Quote: "Allow and log, prompt the user for confirmation, redirect the agent, restrict access, and more. Actions happen in the moment, not after the fact."Source: [island.io](https://www.island.io/product/ai/ai-runtime-protection), read 2026-11-16. - **island-18** (vendor): Island says it reads the prompt, response, every tool call, file access, skill and sub-agent, and classifies prompt injection, jailbreak, agent drift, malicious MCP servers and data leakage inline.Quote: "Real-time classifiers catch known threat types as they happen: prompt injection, jailbreak, agent drift, malicious Model Context Protocol (MCP) servers, data leakage, and unwanted content"Source: [island.io](https://www.island.io/product/ai/ai-runtime-protection), read 2026-11-16. - **island-19** (vendor): Island says a log entry is created for the administrator on every intervention, and audit logs record whether a human or an agent acted (press release dated 2026-10-29).Quote: "Audit logs record whether a human or an agent took the action"Source: [island.io](https://www.island.io/press/island-brings-the-enterprise-browser-to-ai-agents), read 2026-11-16. - **island-20** (vendor): Island says on 2026-08-11 that its Claude Inference Hooks integration needs no new install: server to server, no browser, extension or MDM profile.Quote: "Nothing new to install - no browser, no extension, no MDM profile."Source: [island.io](https://www.island.io/blog/island-integrates-with-claudes-new-inference-hooks-for-inline-ai-security), read 2026-11-16. - **island-21** (vendor): Island says it governs BYOD and unmanaged devices without MDM, VDI or heavy endpoint agents, and that on devices without Island Desktop it collects through a supported EDR platform.Quote: "On devices without Island Desktop, it collects through a supported EDR platform, so the two work together."Source: [island.io](https://www.island.io/product/ai/agentic-endpoint-posture), read 2026-11-16. - **island-22** (none-found): None found on 2026-11-16: looked on the Island AI product pages and sitemap for a time-to-deploy figure, supported operating-system list, inline decision latency and fail-open or fail-closed behavior; the product page only says the classifiers are small models fast enough to sit inline.Source: [island.io](https://www.island.io/product/ai/ai-runtime-protection), read 2026-11-16. - **island-23** (none-found): None found on 2026-11-16: looked for a published price at the request-a-quote page and a public documentation site (docs.island.io returned 404); pricing is by quote.Source: [island.io](https://www.island.io/request-a-quote), read 2026-11-16. - **island-24** (vendor): Island published research on 800 fake AI skills and MCP servers that delivered malware, readable without sign-in, and says it scanned 33,563 public MCP servers.Quote: "Island scanned 33,563 public MCP servers covering 475,865 tools."Source: [island.io](https://www.island.io/product/ai/agentic-endpoint-posture), read 2026-11-16. - **island-25** (vendor): Island says its Enterprise Agentic Control Plane ties Compliance API, endpoint and network evidence into one chain under one identity, and its AI page says audit data is exportable for compliance.Quote: "exportable compliance data that regulated enterprises need across every AI surface"Source: [island.io](https://www.island.io/product/ai), read 2026-11-16. - **island-26** (vendor): Island states in its Gartner Peer Insights Customers' Choice release (2026-10-14) and on its home page that it is a 2026 Customers' Choice for Secure Enterprise Browsers based on 331 verified reviews.Quote: "Island is a Customers' Choice in the 2026 Gartner Peer Insights Voice of the Customer for Secure Enterprise Browsers, based on 331 verified reviews from customers."Source: [island.io](https://www.island.io), read 2026-11-16. - **island-27** (third-party): TechCrunch reported on 2024-04-30 that Island raised $175 million at a $3 billion valuation with Sequoia involved, and had about 200 customers.Quote: "the company has around 200 customers and has been growing steadily over the past couple of years"Source: [techcrunch.com](https://techcrunch.com/2024/04/30/with-175m-in-new-funding-island-wants-the-browser-to-be-at-the-center-of-enterprise-security/), read 2026-11-16. - **island-28** (third-party): SiliconANGLE (2025-05-13) covered Island as an enterprise browser maker built on Chromium, calling it a HyperCUBEd Innovation Startup.Quote: "By building on the open-source Chromium project, Island created a browser that's fully manageable by IT"Source: [siliconangle.com](https://siliconangle.com/2025/05/13/island-technology-enterprise-browser-security-productivity-cubedawards25/), read 2026-11-16. - **island-29** (vendor): Island says two detection engines run in its own infrastructure, one for session patterns and agentic investigators that raise incidents tied to the session and the person.Quote: "Two detection engines run inside Island's own infrastructure."Source: [island.io](https://www.island.io/product/ai/ai-runtime-protection), read 2026-11-16. - **island-30** (none-found): None found on 2026-11-16: looked for named SIEM, SOAR or ticketing export integrations on the AI product pages; only OpenTelemetry ingest, EDR, MDM, Okta and Entra integrations were named.Source: [island.io](https://www.island.io/product/ai/ai-runtime-protection), read 2026-11-16. - **island-31** (vendor): Island says on 2026-10-29 it launched the Enterprise Browser for Agents, which governs data movement, credential use and outputs for endpoint or cloud agents calling the browser.Quote: "Island governs browser activities from the moment an endpoint or cloud agent calls the browser"Source: [island.io](https://www.island.io/press/island-brings-the-enterprise-browser-to-ai-agents), read 2026-11-16. ## News mentioning Island - [Island raises $400 million Series F at a $6.4B valuation](/news/island-400-million-series-f-agentic-control-plane/) ## Related pages - [All vendor profiles](/vendors/) - [CrowdStrike profile](/vendors/crowdstrike/) - [Lasso Security profile](/vendors/lasso-security/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Lasso Security AI agent security URL: https://aiagentsecuritysolutions.com/vendors/lasso-security/ Updated: 2026-11-16 Lasso Security says it discovers agents across Vertex AI, Microsoft Copilot, AWS Bedrock and Salesforce Agentforce through read-only integrations, and governs endpoint coding agents and MCP connections agentlessly, with Claude Code lifecycle hooks via enterprise management. Gateway mode blocks or masks traffic and integrates with Kong, Envoy AI Gateway, LiteLLM and other gateways. The vendor says it blocks or quarantines agent actions and out-of-scope tool calls before execution, streams audit trails to a SIEM and has an EDR integration. A third party reports $37 million raised. Browser-agent control, agent credential revocation, a price and public documentation: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers cloud agent builders, endpoint coding agents, gateways and audit with named integrations (lasso-security-32, lasso-security-16, lasso-security-26). **Weakness:** No public docs or price, and no browser-agent or agent-credential revocation controls found (lasso-security-52, lasso-security-51, lasso-security-35; none found on 2026-11-16). ## Scores and evidence Lasso Security is joint 8th in the overall ranking at 58.4 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Lasso Security: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 6 of 10 | 6 | The vendor inventories named coding agents and MCP connections agentlessly; local enforcement detail and OS coverage are limited. | [lasso-security-13](#fact-lasso-security-13), [lasso-security-19](#fact-lasso-security-19), [lasso-security-26](#fact-lasso-security-26), [lasso-security-27](#fact-lasso-security-27) | | Coverage: browser layer | 2 of 10 | 6 | A usage-control extension exists; browser agent or AI browser coverage none found. | [lasso-security-36](#fact-lasso-security-36), [lasso-security-37](#fact-lasso-security-37) | | Coverage: cloud workspace layer | 6 of 10 | 6 | Discovery across Vertex, Copilot, Bedrock and Agentforce via read-only integrations is described; runtime control in those platforms is limited. | [lasso-security-32](#fact-lasso-security-32), [lasso-security-33](#fact-lasso-security-33), [lasso-security-45](#fact-lasso-security-45) | | Coverage: identity layer | 3 of 10 | 6 | Role-based MCP access and over-privileged agent flags exist; credential inventory and revocation none found. | [lasso-security-24](#fact-lasso-security-24), [lasso-security-34](#fact-lasso-security-34), [lasso-security-35](#fact-lasso-security-35) | | Coverage: gateway layer | 7 of 10 | 6 | Gateway mode blocks or masks traffic and integrates with named AI gateways and a firewall; vendor claims only. | [lasso-security-14](#fact-lasso-security-14), [lasso-security-16](#fact-lasso-security-16), [lasso-security-44](#fact-lasso-security-44) | | Stops the action | 7 of 10 | 20 | The vendor blocks or quarantines agent actions and out-of-scope coding-agent tool calls before execution; vendor claims only. | [lasso-security-10](#fact-lasso-security-10), [lasso-security-12](#fact-lasso-security-12), [lasso-security-14](#fact-lasso-security-14) | | Visibility | 7 of 10 | 15 | Agent discovery, execution graphs, connection maps and tool-call audit trails are described; vendor claims only. | [lasso-security-25](#fact-lasso-security-25), [lasso-security-32](#fact-lasso-security-32), [lasso-security-38](#fact-lasso-security-38), [lasso-security-39](#fact-lasso-security-39) | | Deployment | 7 of 10 | 10 | Agentless, API, gateway and extension modes starting in hours are described; vendor claims only. | [lasso-security-33](#fact-lasso-security-33), [lasso-security-36](#fact-lasso-security-36), [lasso-security-46](#fact-lasso-security-46), [lasso-security-47](#fact-lasso-security-47), [lasso-security-48](#fact-lasso-security-48) | | Policy and governance | 6 of 10 | 10 | Audit trails, framework mapping, SIEM streaming and an EDR integration are described; ticketing and IdP integrations none found. | [lasso-security-11](#fact-lasso-security-11), [lasso-security-25](#fact-lasso-security-25), [lasso-security-40](#fact-lasso-security-40), [lasso-security-43](#fact-lasso-security-43), [lasso-security-42](#fact-lasso-security-42) | | Transparency | 3 of 10 | 10 | Research is published; a price and public documentation none found. | [lasso-security-49](#fact-lasso-security-49), [lasso-security-51](#fact-lasso-security-51), [lasso-security-52](#fact-lasso-security-52) | | Proof | 7 of 10 | 5 | A third party reports $37 million raised and names BMW, eToro and DHS as customers; analyst coverage is vendor-cited. | [lasso-security-02](#fact-lasso-security-02), [lasso-security-03](#fact-lasso-security-03), [lasso-security-04](#fact-lasso-security-04) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | joint 8th | 58.4 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | joint 10th | 62.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 10th | 50.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | joint 5th | 62.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 11th | 53.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 3rd | 65.0 | ## Product, deployment, price and owner - Product assessed: Lasso AI Security Platform: AI Agents Security, Discovery & AI-BOM, AI Usage Control, AI Gateway Security - Deployment: Agentless (read-only CI and cloud integrations), API, SDK, gateway or firewall integration, browser extension, and Claude Code lifecycle hooks via enterprise management - Time to deploy: Lasso says start in hours, not weeks, and usage control in minutes - Price: none found on 2026-11-16 - Owner: Lasso Security Ltd. - Website: [lasso.security](https://www.lasso.security) ## Facts and sources - **lasso-security-01** (vendor): Lasso says its AI Security Platform gives enterprises visibility, control and protection across AI models, agents and apps.Quote: "Lasso's AI Security Platform gives enterprises visibility, control, and protection across AI models, agents, and apps."Source: [lasso.security](https://www.lasso.security/), read 2026-11-16. - **lasso-security-02** (third-party): SiliconANGLE reports that Lasso Security Ltd. was founded in 2023 by Schulman and Dror and has raised more than $37 million in total, including a $30 million round announced on 2026-09-02 led by ClearSky Advisors.Quote: "Investors have now put more than $37 million into the company."Source: [siliconangle.com](https://siliconangle.com/2026/09/02/lasso-security-launches-cpu-only-ai-guardrail-raises-30m/), read 2026-11-16. - **lasso-security-03** (third-party): SiliconANGLE names BMW, eToro, Fiverr and the U.S. Department of Homeland Security as Lasso customers (reported 2026-09-02).Quote: "BMW AG is a customer, and so are eToro Group Ltd., Fiverr International Ltd. and the U.S. Department of Homeland Security."Source: [siliconangle.com](https://siliconangle.com/2026/09/02/lasso-security-launches-cpu-only-ai-guardrail-raises-30m/), read 2026-11-16. - **lasso-security-04** (vendor): On 2026-09-15 Lasso said Latio Tech's 2026 AI Security Market Report named it a platform leader (vendor's own account of a third-party report).Quote: "In the report, Lasso is named as one of the major Platform Leaders."Source: [lasso.security](https://www.lasso.security/blog/lasso-named-a-leader-in-latios-2026-ai-security-market-report), read 2026-11-16. - **lasso-security-05** (third-party): On 2026-09-01 TechCentral (a promoted article by NEWORDER) said NEWORDER is Lasso's official partner in South Africa and that Lasso is certified to ISO 27001 and SOC 2.Quote: "Lasso Security was founded in Tel Aviv in 2023, is certified to ISO 27001 and SOC 2, and has customers across Europe, the US and Israel."Source: [techcentral.co.za](https://techcentral.co.za/neworder-lasso-security-ai-agent-controls/285558/), read 2026-11-16. - **lasso-security-10** (vendor): Lasso says it enforces policy inline so a harmful agent action can be stopped before it executes, and that it can block or quarantine an action.Quote: "preventing unauthorized actions requires inline enforcement that can stop a harmful action before it executes and limit what each agent is permitted to do."Source: [lasso.security](https://www.lasso.security/platform/ai-detection-response), read 2026-11-16. - **lasso-security-11** (vendor): Lasso says on detection it can block or quarantine an action before it executes, alert with trace context, and feed findings to SIEM.Quote: "Lasso can block or quarantine an action before it executes, alert security teams with full trace context for investigation, and feed findings into your existing SIEM and observability workflows."Source: [lasso.security](https://www.lasso.security/platform/ai-detection-response), read 2026-11-16. - **lasso-security-12** (vendor): Lasso says it inspects every coding-agent tool call before execution and flags or blocks anything outside the permitted scope.Quote: "Lasso inspects every tool call before it is executed."Source: [lasso.security](https://www.lasso.security/use-cases/ai-coding-assistants), read 2026-11-16. - **lasso-security-13** (vendor): Lasso says its coding-assistant coverage addresses agents that read codebases, execute shell commands and call MCP servers (Claude Code, Cursor, Codex, OpenCode).Quote: "Claude Code, Cursor, and Codex read your codebase, executes shell commands, calls MCPs, and takes actions on your behalf."Source: [lasso.security](https://www.lasso.security/use-cases/ai-coding-assistants), read 2026-11-16. - **lasso-security-14** (vendor): Lasso says gateway mode can block an interaction before it reaches the model or before a response returns, mask sensitive data in transit, or allow and alert.Quote: "block the interaction before it reaches the model or before a response is returned, mask sensitive data in transit, or allow and alert where interruption would affect a critical workflow"Source: [lasso.security](https://www.lasso.security/use-cases/ai-gateway-security), read 2026-11-16. - **lasso-security-16** (vendor): Lasso says it integrates with Kong, Envoy AI Gateway, TrueFoundry, Portkey, LiteLLM and other AI gateways without source code changes.Quote: "Lasso integrates with Kong, Envoy AI Gateway, TrueFoundry, Portkey, LiteLLM, and others."Source: [lasso.security](https://www.lasso.security/use-cases/ai-gateway-security), read 2026-11-16. - **lasso-security-18** (vendor): Lasso says it blocks injected or guardrail-bypassing inputs and decodes over 3,000 obfuscation techniques at the intent layer.Quote: "Lasso decodes over 3,000 obfuscation techniques and detects threats at the intent layer. It blocks inputs designed to bypass safety guardrails or trick agents into performing unauthorized actions"Source: [lasso.security](https://www.lasso.security/platform/ai-agents-security), read 2026-11-16. - **lasso-security-19** (vendor): Lasso says it discovers MCP servers across the organization and inventories Claude Code, Claude Desktop, Cursor, Windsurf and custom agent connections.Quote: "Automatically discover all MCP servers connected across your organization. Inventory Claude Code and Desktop, Cursor, Windsurf, and custom agent connections."Source: [lasso.security](https://www.lasso.security/use-cases/mcp-security), read 2026-11-16. - **lasso-security-24** (vendor): Lasso says it enforces role-based control over which users and teams can connect to which MCP servers.Quote: "Control which users and teams can connect to which MCP servers. Enforce role-based permissions that allow approved use cases while blocking unauthorized connections."Source: [lasso.security](https://www.lasso.security/use-cases/mcp-security), read 2026-11-16. - **lasso-security-25** (vendor): Lasso says it keeps audit trails of every MCP connection and tool call and exports logs to a SIEM.Quote: "Maintain complete audit trails of every MCP connection and tool call. Export logs to your SIEM."Source: [lasso.security](https://www.lasso.security/use-cases/mcp-security), read 2026-11-16. - **lasso-security-26** (vendor): Lasso says it governs AI activity on employee endpoints agentlessly, so visibility into AI tools, coding agents and MCP connections starts on day one.Quote: "Lasso governs AI activity on employee endpoints through an agentless approach, so security teams get visibility into AI tools, coding agents, and MCP connections from day one"Source: [lasso.security](https://www.lasso.security/blog/securing-ai-on-endpoints-how-to-govern-ai-tools-coding-agents-and-mcp-connections), read 2026-11-16. - **lasso-security-27** (vendor): Lasso says teams with EDR or MDM tooling can connect Lasso to that stack to extend endpoint coverage.Quote: "Teams that already run EDR or MDM tooling can connect Lasso to that existing stack, extending endpoint coverage through infrastructure they already maintain"Source: [lasso.security](https://www.lasso.security/blog/securing-ai-on-endpoints-how-to-govern-ai-tools-coding-agents-and-mcp-connections), read 2026-11-16. - **lasso-security-30** (none-found): None found on 2026-11-16: supported operating systems and a time to deploy for MCP discovery on developer machines, searched on the MCP security, coding assistants and endpoint pages.Source: [lasso.security](https://www.lasso.security/use-cases/mcp-security), read 2026-11-16. - **lasso-security-31** (observed): Lasso publishes an open-source MCP gateway (MIT license) that intercepts requests and responses to sanitize sensitive information via plugins.Quote: "A plugin-based gateway that orchestrates other MCPs and allows developers to build upon it enterprise-grade agents."Source: [github.com](https://github.com/lasso-security/mcp-gateway), read 2026-11-16. - **lasso-security-32** (vendor): Lasso says it discovers agents across Vertex AI, Microsoft Copilot, AWS Bedrock, Salesforce Agentforce and custom-built agents.Quote: "Lasso provides platform-agnostic discovery across Vertex AI, Microsoft Copilot, AWS Bedrock, Salesforce Agentforce, and custom-built agents."Source: [lasso.security](https://www.lasso.security/platform/ai-agents-security), read 2026-11-16. - **lasso-security-33** (vendor): Lasso says its discovery is agentless and read-only, using CI pipelines and native cloud and agent-builder integrations.Quote: "Agentless, read-only integration into CI pipelines and native integrations with cloud providers and third-party agent builder tools."Source: [lasso.security](https://www.lasso.security/platform/discovery-ai-bom), read 2026-11-16. - **lasso-security-34** (vendor): Lasso says it identifies and categorizes overprivileged or high-risk agents.Quote: "Identify and categorize overprivileged or high-risk agents to stop dangerous usage patterns and reduce the attack surface."Source: [lasso.security](https://www.lasso.security/platform/ai-agents-security), read 2026-11-16. - **lasso-security-35** (none-found): None found on 2026-11-16: agent credential or token inventory and permission revocation, searched on the AI Agents Security and AI Usage Control pages.Source: [lasso.security](https://www.lasso.security/platform/ai-agents-security), read 2026-11-16. - **lasso-security-36** (vendor): Lasso says AI usage control deploys through a browser extension, proxy integrations and data integrations, and can be running in minutes.Quote: "Enable full AI usage control in minutes through Lasso’s lightweight deployment paths like our browser extension, proxy integrations, and data integrations."Source: [lasso.security](https://www.lasso.security/platform/ai-usage-control), read 2026-11-16. - **lasso-security-37** (none-found): None found on 2026-11-16: documented coverage of browser agents or AI browsers as a product capability (only a blog on AI browser risks), searched on the AI Usage Control page and the AI browsers blog.Source: [lasso.security](https://www.lasso.security/blog/the-rise-of-ai-browsers), read 2026-11-16. - **lasso-security-38** (vendor): Lasso says it maps tool calls, RAG retrievals and sub-agent communications into an execution graph.Quote: "Lasso monitors every interaction across your AI stack, from tool calls and RAG retrievals to sub-agent communications."Source: [lasso.security](https://www.lasso.security/platform/ai-detection-response), read 2026-11-16. - **lasso-security-39** (vendor): Lasso says it maps agents' connections to third-party tools, databases and internal services to show attack paths.Quote: "Map your agents and visualize their connections to third-party tools, databases, and internal services to identify potential attack paths for AI threats."Source: [lasso.security](https://www.lasso.security/platform/ai-agents-security), read 2026-11-16. - **lasso-security-40** (vendor): Lasso says exportable audit trails are updated on every interaction and mapped to NIST AI RMF, OWASP, EU AI Act, ISO 42001 and others.Quote: "Exportable audit trails are updated on every interaction, not generated on demand when an audit request arrives."Source: [lasso.security](https://www.lasso.security/use-cases/ai-gateway-security), read 2026-11-16. - **lasso-security-41** (vendor): Lasso says alerts and audit logs can be viewed in Lasso or streamed to a SIEM.Quote: "View alerts and audit logs directly in Lasso, or stream findings into your SIEM to investigate agent incidents alongside the rest of your security telemetry."Source: [lasso.security](https://www.lasso.security/platform/ai-detection-response), read 2026-11-16. - **lasso-security-42** (none-found): None found on 2026-11-16: documented ticketing or identity-provider integrations, or SOAR export, searched on the detection and response, usage control and MCP pages.Source: [lasso.security](https://www.lasso.security/platform/ai-detection-response), read 2026-11-16. - **lasso-security-43** (vendor): Lasso says its CrowdStrike Falcon integration ingests network signals and combines them with Lasso web and desktop discovery (blog dated 2026-01-26).Quote: "Lasso ingests network signals from CrowdStrike Falcon and combines them with Lasso’s web and desktop discovery."Source: [lasso.security](https://www.lasso.security/blog/ai-agents-discovery-lasso-crowdstrike-falcon), read 2026-11-16. - **lasso-security-44** (vendor): Lasso says it integrates with Palo Alto Networks Next-Generation Firewall to extend monitoring and runtime protection from the browser to the network (blog dated 2026-02-18).Quote: "Lasso Security integrates directly with Palo Alto Networks’ Next-Generation Firewall infrastructure, extending AI monitoring, risk management, and runtime protection from the browser to the network level."Source: [lasso.security](https://www.lasso.security/blog/securing-desktop-ai-agents-palo-alto-ngfw-integration), read 2026-11-16. - **lasso-security-45** (vendor): On 2026-09-29 Lasso said its Claude Compliance API integration is generally available, bringing Claude Enterprise users, MCP servers and skills into its inventory.Quote: "Today we're launching Lasso's integration with Claude's Compliance API, and it is generally available starting now."Source: [lasso.security](https://www.lasso.security/blog/bringing-claude-enterprise-activity-into-lasso-with-claudes-compliance-api), read 2026-11-16. - **lasso-security-46** (vendor): Lasso says it deploys via API, firewall or AI gateway integration with no architectural changes.Quote: "Deploy via API, Firewall, or AI Gateway Integration"Source: [lasso.security](https://www.lasso.security/platform/ai-detection-response), read 2026-11-16. - **lasso-security-47** (vendor): Lasso says it runs from the Lasso tenant and starts in hours, not weeks, with no agents or code changes.Quote: "No agents. No code changes. No source code access. Run from the Lasso tenant and start in hours, not weeks."Source: [lasso.security](https://www.lasso.security/platform/ai-detection-response), read 2026-11-16. - **lasso-security-48** (vendor): Lasso says it deploys with a single line of code via Gateway, API or SDK and ships policy templates aligned to OWASP, MITRE and NIST.Quote: "Lasso deploys with a single line of code via Gateway, API, or SDK."Source: [lasso.security](https://www.lasso.security/platform/ai-agents-security), read 2026-11-16. - **lasso-security-49** (vendor): On 2026-09-17 Lasso published research on how LLM text watermarking changes refusals and agent tool calling.Quote: "We find that it does, in both model refusal behavior and agent tool calling."Source: [lasso.security](https://www.lasso.security/blog/the-provenance-tax-understanding-the-impact-of-llm-watermarking-on-ai-agent-behavior), read 2026-11-16. - **lasso-security-50** (vendor): On 2026-09-09 Lasso published a technical post on LEAP, its CPU-based guardrail, with benchmark claims.Quote: "Introducing LEAP: CPU-based AI Security Guardrails with GPU-Class Accuracy"Source: [lasso.security](https://www.lasso.security/blog/introducing-leap-cpu-based-ai-security-guardrails-with-gpu-class-accuracy), read 2026-11-16. - **lasso-security-51** (none-found): None found on 2026-11-16: a published price; the pricing URL returns a 404 page and the site routes to Book a Demo.Source: [lasso.security](https://www.lasso.security/pricing), read 2026-11-16. - **lasso-security-52** (none-found): None found on 2026-11-16: public product documentation; docs.lasso.security loads only a sign-in dashboard titled Lasso Security - Dashboard.Source: [docs.lasso.security](https://docs.lasso.security), read 2026-11-16. - **lasso-security-54** (vendor): Lasso says its AI-SPM defines how agents should behave, which tools they can use and which resources they may access, aligned to OWASP, MITRE and NIST.Quote: "Implement secure-by-design policies that align with OWASP, MITRE, NIST and more and that define how agents should behave, what tools they can use"Source: [lasso.security](https://www.lasso.security/platform/ai-agents-security), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Island profile](/vendors/island/) - [Microsoft profile](/vendors/microsoft/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Microsoft AI agent security URL: https://aiagentsecuritysolutions.com/vendors/microsoft/ Updated: 2026-11-16 Microsoft ties agent security to products many buyers already run. Per its documentation, the Agent 365 registry lists agents with owner, usage and risk, Entra gives agents identities governed by Conditional Access, and Defender for Endpoint discovers local agents on Windows and macOS, a feature the release notes describe as generally available. Purview adds audit records. Agent 365 is listed at $15.00 per user per month, billed yearly, which is the only published price in this set beyond a free edition. Some runtime blocking is in preview, and several controls need Agent 365 or E7 licenses. Dedicated browser-agent control: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** One control plane ties an agent registry, Entra agent identities with Conditional Access, Defender runtime protection and Purview audit, at $15 per user per month (microsoft-36, microsoft-34, microsoft-29, microsoft-49). **Weakness:** No dedicated browser-agent control was found, and several controls are preview or need E7 or Agent 365 licenses (microsoft-40, microsoft-35). ## Scores and evidence Microsoft is 3rd in the overall ranking at 63.0 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Microsoft: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 8 of 10 | 6 | A support matrix of named local agents with MCP data is documented and discovery is confirmed by two third-party reports; runtime blocking is public preview only. | [microsoft-11](#fact-microsoft-11), [microsoft-14](#fact-microsoft-14), [microsoft-19](#fact-microsoft-19), [microsoft-54](#fact-microsoft-54), [microsoft-65](#fact-microsoft-65) | | Coverage: browser layer | 2 of 10 | 6 | Only a general DLP browser control; AI browser or agent extension control none found. | [microsoft-40](#fact-microsoft-40), [microsoft-67](#fact-microsoft-67) | | Coverage: cloud workspace layer | 7 of 10 | 6 | Inventory and real-time protection for platform agents and Bedrock and Vertex workloads are documented; third-party confirmation none found. | [microsoft-29](#fact-microsoft-29), [microsoft-30](#fact-microsoft-30), [microsoft-31](#fact-microsoft-31), [microsoft-32](#fact-microsoft-32) | | Coverage: identity layer | 7 of 10 | 6 | Agent identity accounts, conditional access for agents with stated license needs, and a registry that can block a risky agent are documented; vendor documentation only for available features. | [microsoft-33](#fact-microsoft-33), [microsoft-34](#fact-microsoft-34), [microsoft-35](#fact-microsoft-35), [microsoft-36](#fact-microsoft-36) | | Coverage: gateway layer | 6 of 10 | 6 | An open-source MCP gateway is observed and admins can control tool connections; a productized gateway is limited. | [microsoft-39](#fact-microsoft-39), [microsoft-68](#fact-microsoft-68) | | Stops the action | 6 of 10 | 20 | Tool invocations are evaluated against policy and conditional access can block agent access to resources and MCP servers; pre-execution runtime blocking and document-attack scanning are preview. | [microsoft-19](#fact-microsoft-19), [microsoft-30](#fact-microsoft-30), [microsoft-34](#fact-microsoft-34), [microsoft-43](#fact-microsoft-43) | | Visibility | 7 of 10 | 15 | An agent registry with owner, risk, MCP servers and declared tools plus activity records is documented; vendor documentation only. | [microsoft-17](#fact-microsoft-17), [microsoft-31](#fact-microsoft-31), [microsoft-36](#fact-microsoft-36), [microsoft-37](#fact-microsoft-37) | | Deployment | 7 of 10 | 10 | Onboarding, license requirements, OS list and a 24-hour data collection time are published; vendor documentation only. | [microsoft-12](#fact-microsoft-12), [microsoft-35](#fact-microsoft-35), [microsoft-38](#fact-microsoft-38), [microsoft-61](#fact-microsoft-61) | | Policy and governance | 6 of 10 | 10 | Conditional access, approval workflows, audit records and hunting queries are documented; named SIEM and ticketing exports none found. | [microsoft-17](#fact-microsoft-17), [microsoft-34](#fact-microsoft-34), [microsoft-36](#fact-microsoft-36), [microsoft-37](#fact-microsoft-37), [microsoft-69](#fact-microsoft-69) | | Transparency | 7 of 10 | 10 | Documentation is public and a per-user price is published; research specific to agent security none found. | [microsoft-49](#fact-microsoft-49), [microsoft-52](#fact-microsoft-52), [microsoft-66](#fact-microsoft-66) | | Proof | 5 of 10 | 5 | Third parties cover the launch by a large company; named customers, funding and analyst coverage for this line none found. | [microsoft-56](#fact-microsoft-56), [microsoft-64](#fact-microsoft-64), [microsoft-57](#fact-microsoft-57) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 3rd | 63.0 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 2nd | 69.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 9th | 51.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 4th | 66.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 2nd | 66.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 5th | 63.0 | ## Product, deployment, price and owner - Product assessed: Microsoft Agent 365 with Defender for Endpoint and Defender AI agent protection, Microsoft Entra Agent ID and Conditional Access for agents, Purview - Deployment: Cloud-managed Defender for Endpoint client onboarding (Intune or PowerShell for runtime protection); API-based Content Safety; Purview browser extension; Entra and Defender portal controls - Time to deploy: none found on 2026-11-16 (discovery starts automatically after device onboarding; Purview policies need at least 24 hours) - Price: Agent 365 is listed at $15.00 per user per month, paid yearly; Content Safety Standard shows a dash; Defender for Endpoint and Cloud Apps prices none found on 2026-11-16 on readable pages. - Owner: Microsoft Corporation - Website: learn.microsoft.com ## Facts and sources - **microsoft-08** (vendor): Microsoft says third-party AI site insights in DSPM for AI need the Purview browser extension and onboarded devices as prerequisites.Quote: "Install Microsoft Purview browser extension A prerequisite for third-party AI sites."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/purview/dspm-for-ai), read 2026-11-16. - **microsoft-10** (vendor): Microsoft documents Endpoint DLP detection of pasting content to restricted service domains in supported browsers, evaluated on the pasted content.Quote: "Detects when a user attempts to paste content to a restricted service domain. Evaluation is performed on the content that is being pasted."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/purview/endpoint-dlp-learn-about), read 2026-11-16. - **microsoft-11** (vendor): Microsoft says Defender for Endpoint discovers supported local AI agents on Windows devices, including coding agents, IDE extensions, desktop assistants and agent platforms, and lists them in an AI agent inventory.The exact quote is on the source page.Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint), read 2026-11-16. - **microsoft-12** (vendor): Microsoft says local agent discovery needs no extra deployment beyond onboarding to Defender for Endpoint and starts automatically; Plan 2 is the minimum license.Quote: "You don't need additional deployment, configuration, or scripts beyond the device onboarding requirements. If the device meets all of the prerequisites, agent discovery begins automatically."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/discover-local-ai-agents), read 2026-11-16. - **microsoft-13** (vendor): Microsoft lists Windows, macOS, Linux, Android and iOS as Defender for Endpoint operating systems; local agent discovery is documented for Windows and macOS.Quote: "Microsoft Defender for Endpoint supports the following operating systems: Windows, macOS, Linux, Android, and iOS."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/microsoft-defender-endpoint), read 2026-11-16. - **microsoft-14** (vendor): Microsoft's support matrix names the local agents discovered, such as Claude Code, Claude Desktop, Cursor, Codex CLI, Gemini CLI, GitHub Copilot, Windsurf and Kiro, and marks which also report MCP server configuration.Quote: "Discovers MCP column indicates that Defender reports remote MCP server data in McpServers and local MCP server data in RawAgentInfo.localAgentMetadata.localMcps"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/ai-agent-support-matrix), read 2026-11-16. - **microsoft-17** (vendor): Microsoft says local agent data can be queried in advanced hunting, including an AgentsInfo table that carries MCP servers and declared tools.Quote: "Use advanced hunting to proactively investigate local AI agent presence, understand how agents are configured, and identify the agents and users that carry the most risk."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/discover-local-ai-agents), read 2026-11-16. - **microsoft-19** (vendor): Microsoft says Defender for Endpoint runtime protection inspects the agent loop on endpoints and can block risky activity before it executes; this is a public preview feature.Quote: "Microsoft Defender inspects the agent loop (user prompts, tool calls, and tool responses) and can block risky activity before it executes"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint), read 2026-11-16. - **microsoft-25** (vendor): Microsoft says runtime protection is configured in audit or block mode through local PowerShell or organization-wide Intune.Quote: "For local PowerShell and organization-wide Microsoft Intune configuration options"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/ai-agent-runtime-protection-overview), read 2026-11-16. - **microsoft-28** (vendor): Microsoft Defender integrates with Agent 365 so that real-time protection evaluates tool calls and responses for Work IQ MCP and customer MCP tools onboarded to Agent 365; the default rule audits and custom rules can block before execution.Quote: "custom rules can block matching actions before they execute"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-cloud-apps/ai-agent-protection), read 2026-11-16. - **microsoft-29** (vendor): Microsoft says Defender integrated with Agent 365 gives discovery, posture management, real-time protection and investigation for managed agents, including local agents on supported endpoints.Quote: "Microsoft Defender integrates with Agent 365 to secure all managed agents, including local AI agents on supported endpoints"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-cloud-apps/ai-agent-protection), read 2026-11-16. - **microsoft-30** (vendor): Microsoft says real-time protection evaluates tool invocations against policy and near-real-time detections cover indirect prompt injection, secret leakage and suspicious IP access.Quote: "Near-real-time detections analyze Agent 365 observability data for threats such as indirect prompt injection, evasion techniques, malicious content propagation, secret leakage"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-cloud-apps/ai-agent-protection), read 2026-11-16. - **microsoft-31** (vendor): Microsoft says the Defender inventory covers agents built with Copilot Studio, Foundry, Microsoft 365 and supported non-Microsoft platforms, plus local agents.Quote: "Agents built with Microsoft Copilot Studio, Microsoft Foundry, Microsoft 365, and supported non-Microsoft platforms ."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-xdr/security-for-ai/ai-agent-inventory), read 2026-11-16. - **microsoft-32** (vendor): Microsoft says Defender for Cloud discovers AI workloads on Azure OpenAI, Azure AI Foundry, Azure Machine Learning, Amazon Bedrock and Google Vertex AI, and that agent-level discovery moved to Agent 365 licensing effective July 1, 2026.Quote: "Effective July 1, 2026, AI agent discovery and security posture for Microsoft Foundry agents and third-party cloud agents require a Microsoft Agent 365 license."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/azure/defender-for-cloud/ai-security-posture), read 2026-11-16. - **microsoft-33** (vendor): Microsoft says agent identities are Entra ID accounts that give AI agents unique identification and authentication.Quote: "Agent identities are identity accounts within Microsoft Entra ID that provide unique identification and authentication capabilities for AI agents."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/entra/agent-id/identity-platform/what-is-agent-id), read 2026-11-16. - **microsoft-34** (vendor): Microsoft says Conditional Access for agents evaluates agent, user, device and risk signals to allow, block or limit access, including access to MCP servers.Quote: "determine when to allow, block, or limit access, or require more verification steps"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/entra/identity/conditional-access/agent-id), read 2026-11-16. - **microsoft-35** (vendor): Microsoft says Conditional Access for agents requires Microsoft 365 E7 or an Agent 365 license paired with at least Entra P1 or Microsoft 365 E3.Quote: "Microsoft Agent 365 license paired with at least Microsoft Entra P1 or Microsoft 365 E3."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/entra/identity/conditional-access/agent-id), read 2026-11-16. - **microsoft-36** (vendor): Microsoft says the Agent 365 registry lists agents with owner, usage and risk, shows risk signals from Defender, Entra and Purview, and lets admins block a risky agent right away.Quote: "The agent registry shows risk signals from Microsoft Defender, Microsoft Entra, and Microsoft Purview in one place."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/microsoft-agent-365/overview), read 2026-11-16. - **microsoft-37** (vendor): Microsoft says Purview keeps a record of agent activity for audits and admins control which tools agents can connect to.Quote: "Microsoft Purview keeps a record of agent activity, so you're ready for audits."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/microsoft-agent-365/overview), read 2026-11-16. - **microsoft-38** (vendor): Microsoft says Agent 365 is generally available for the Commercial segment per user, works best with Microsoft E5, and needs at least one licensed user.Quote: "Microsoft Agent 365 is generally available for the Commercial segment on a per user basis."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/microsoft-agent-365/overview), read 2026-11-16. - **microsoft-39** (observed): Microsoft publishes an open-source MCP Gateway, a reverse proxy and management layer for MCP servers on Kubernetes with authorization and lifecycle management (an MIT-licensed GitHub repo, last pushed 2026-11-13).Quote: "reverse proxy and management layer for [Model Context Protocol (MCP)](https://modelcontextprotocol.io/introduction) servers"Source: [github.com](https://github.com/microsoft/mcp-gateway), read 2026-11-16. - **microsoft-40** (none-found): None found on 2026-11-16: no Microsoft page read describes a dedicated AI browser or AI extension agent control in Defender; the browser-side evidence is the Purview browser extension and Endpoint DLP paste control.Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/purview/dspm-for-ai), read 2026-11-16. - **microsoft-43** (vendor): Microsoft says Prompt Shields can scan document attacks at the tool response point for agents on supported tools (Preview) and that document attack detection is off by default.Quote: "For agents, Prompt Shields can also scan document attacks at the tool response intervention point for supported tools (Preview)."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/azure/ai-services/content-safety/concepts/jailbreak-detection), read 2026-11-16. - **microsoft-49** (vendor): Microsoft lists Agent 365 at $15.00 per user per month, paid yearly, on its product page.Quote: "$15.00 user/month, paid yearly (Annual commitment)"Source: [microsoft.com](https://www.microsoft.com/en-us/microsoft-agent-365), read 2026-11-16. - **microsoft-50** (vendor): Microsoft's Azure AI Content Safety pricing page shows a free tier of 5,000 text records per month and a Standard tier whose price displays as a dash.The exact quote is on the source page.Source: [azure.microsoft.com](https://azure.microsoft.com/en-us/pricing/details/content-safety/), read 2026-11-16. - **microsoft-51** (none-found): None found on 2026-11-16: no price for Defender for Cloud Apps or Defender for Endpoint standalone is readable on the learn pages; they point to Microsoft 365 E5 inclusion and separate pricing pages that returned 403 to a script.Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/microsoft-defender-endpoint), read 2026-11-16. - **microsoft-52** (vendor): Microsoft documentation on learn.microsoft.com is readable without sign-in, with last-updated dates (Agent 365 overview 2026-11-12).Quote: "Last updated on 2026-11-12"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/microsoft-agent-365/overview), read 2026-11-16. - **microsoft-54** (third-party): On 2026-10-20, Help Net Security reported Microsoft is expanding discovery to 35 more local agent types for a total of 60, including Claude Code, OpenAI Codex and GitHub Copilot.Quote: "Microsoft Security is expanding discovery to 35 additional types of local agents, bringing the total supported to 60"Source: [helpnetsecurity.com](https://www.helpnetsecurity.com/2026/10/20/microsofts-ai-agents-threats-guardian-agents/), read 2026-11-16. - **microsoft-55** (third-party): On 2026-10-20, Help Net Security reported Microsoft announced Dynamic Permissions giving agents only the authority needed for an approved task, with higher-impact actions able to require verified human approval.Quote: "Higher-impact actions can require verified human approval."Source: [helpnetsecurity.com](https://www.helpnetsecurity.com/2026/10/20/microsofts-ai-agents-threats-guardian-agents/), read 2026-11-16. - **microsoft-56** (third-party): On 2026-10-20, Help Net Security quoted Satya Nadella, CEO of Microsoft, on treating frontier models as insider risks, and said Project Perception shares signals with Agent 365.Quote: "Satya Nadella , CEO of Microsoft, wrote"Source: [helpnetsecurity.com](https://www.helpnetsecurity.com/2026/10/20/microsofts-ai-agents-threats-guardian-agents/), read 2026-11-16. - **microsoft-57** (none-found): None found on 2026-11-16: named customers, funding or analyst coverage for Microsoft's agent security line could not be read from the pages fetched (the Gartner and customer story items seen only as news headlines were not opened).Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/microsoft-agent-365/overview), read 2026-11-16. - **microsoft-59** (vendor): Microsoft owns and sells these products itself: Defender for Endpoint is described as the endpoint security pillar of Microsoft Defender.Quote: "As the endpoint security pillar of Microsoft Defender , Defender for Endpoint feeds endpoint signals into the unified Defender portal."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/defender-endpoint/microsoft-defender-endpoint), read 2026-11-16. - **microsoft-61** (vendor): Microsoft Purview says DSPM for AI policies need at least 24 hours to collect data before results appear.Quote: "Allow at least 24 hours for these new policies to collect data to display the results in Data Security Posture Management for AI"Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/purview/dspm-for-ai), read 2026-11-16. - **microsoft-62** (third-party): Microsoft says it is extending Entra Agent ID with identity types for local agents and previously unknown agents found through security sensors.Quote: "extending Microsoft Entra Agent ID with identity types for additional agent patterns, including local agents"Source: [helpnetsecurity.com](https://www.helpnetsecurity.com/2026/10/20/microsofts-ai-agents-threats-guardian-agents/), read 2026-11-16. - **microsoft-63** (third-party): On 2026-10-19, SiliconANGLE reported Microsoft extended Conditional Access to agents so that rising risk can hold higher-impact actions for human approval and, at the top of the scale, isolate or stop the runtime.Quote: "Higher-impact actions can be held for verified human approval, and at the top of the scale the runtime is isolated or stopped entirely."Source: [siliconangle.com](https://siliconangle.com/2026/10/19/microsoft-targets-the-agentic-insider-with-new-controls-for-rogue-ai-agents/), read 2026-11-16. - **microsoft-64** (third-party): On 2026-10-19, SiliconANGLE covered the launch and named Satya Nadella as Microsoft Chairman and Chief Executive and Hayete Gallot as executive vice president of Microsoft Security.Quote: "Microsoft Chairman and Chief Executive Satya Nadella laid out the thinking behind the controls in an Oct. 10 essay"Source: [siliconangle.com](https://siliconangle.com/2026/10/19/microsoft-targets-the-agentic-insider-with-new-controls-for-rogue-ai-agents/), read 2026-11-16. - **microsoft-65** (third-party): On 2026-10-19, SiliconANGLE reported Microsoft's discovery tools can now spot 60 types of local agents on employee machines after 35 were added.Quote: "discovery tools can now spot 60 types of local agents on employee machines after 35 were added."Source: [siliconangle.com](https://siliconangle.com/2026/10/19/microsoft-targets-the-agentic-insider-with-new-controls-for-rogue-ai-agents/), read 2026-11-16. - **microsoft-66** (third-party): On 2026-10-19, SiliconANGLE reported Microsoft gave no availability dates or pricing for the new agent controls announced that day.Quote: "Microsoft did not give availability dates or pricing for the new controls"Source: [siliconangle.com](https://siliconangle.com/2026/10/19/microsoft-targets-the-agentic-insider-with-new-controls-for-rogue-ai-agents/), read 2026-11-16. - **microsoft-67** (vendor): Microsoft says Endpoint DLP blocks an upload from an unallowed browser and redirects the user to Microsoft Edge, a browser-side control that is not specific to AI agents.Quote: "If they're using an unallowed browser, the upload activity is blocked and the user is redirected to use Microsoft Edge."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/purview/endpoint-dlp-learn-about), read 2026-11-16. - **microsoft-68** (vendor): Microsoft says Agent 365 lets admins control which tools agents can connect to.Quote: "You control which tools agents can connect to."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/microsoft-agent-365/overview), read 2026-11-16. - **microsoft-69** (vendor): Microsoft says admins review and approve agents before publishing them to the organization and decide who can use them.Quote: "You review and approve agents before you publish them to your organization, and you decide who can use them."Source: [learn.microsoft.com](https://learn.microsoft.com/en-us/microsoft-agent-365/overview), read 2026-11-16. ## News mentioning Microsoft - [Microsoft adds agent discovery and Dynamic Permissions](/news/microsoft-agentic-insider-controls-dynamic-permissions/) ## Related pages - [All vendor profiles](/vendors/) - [Lasso Security profile](/vendors/lasso-security/) - [Noma Security profile](/vendors/noma-security/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Noma Security AI agent security URL: https://aiagentsecuritysolutions.com/vendors/noma-security/ Updated: 2026-11-16 Noma Security names Microsoft Copilot Studio, Salesforce Agentforce, ServiceNow, Bedrock and Foundry among the agent platforms it covers, with runtime action evaluation. On 28 Sep 2026 it announced endpoint agent security with enforcement across Claude Code and other named coding agents, using discovery through existing EDR or MDM. Agents get attributable identities tied to builders, inherited permissions and identity provider groups, and policy can block, mask, steer or kill an action, though available enforcement depends on the integration. Third parties report $132 million raised. Browser-layer control, its own gateway, a price and public documentation: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers endpoint, SaaS and cloud agents with one policy plane and named platforms (noma-security-09, noma-security-12, noma-security-14). **Weakness:** No price, no public docs and no browser-layer coverage found on 2026-11-16 (noma-security-42, noma-security-43, noma-security-10). ## Scores and evidence Noma Security is 13th in the overall ranking at 54.6 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Noma Security: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 6 of 10 | 6 | Discovery via existing EDR or MDM and enforcement across named coding agents are claimed; enforcement mechanism and OS detail are limited. | [noma-security-08](#fact-noma-security-08), [noma-security-09](#fact-noma-security-09) | | Coverage: browser layer | 1 of 10 | 6 | Only a mention of an agent that works with browsers; browser-layer control none found. | [noma-security-10](#fact-noma-security-10), [noma-security-11](#fact-noma-security-11) | | Coverage: cloud workspace layer | 7 of 10 | 6 | Copilot Studio, Agentforce, ServiceNow, Bedrock and Foundry are named with runtime action evaluation; vendor claims only. | [noma-security-12](#fact-noma-security-12), [noma-security-13](#fact-noma-security-13), [noma-security-14](#fact-noma-security-14), [noma-security-41](#fact-noma-security-41) | | Coverage: identity layer | 6 of 10 | 6 | Agents get attributable identities tied to builders, inherited permissions and IdP groups; revocation none found. | [noma-security-15](#fact-noma-security-15), [noma-security-16](#fact-noma-security-16), [noma-security-17](#fact-noma-security-17) | | Coverage: gateway layer | 6 of 10 | 6 | Policy is enforced through existing AI and MCP gateways, hooks and SDKs; the vendor's own gateway none found. | [noma-security-18](#fact-noma-security-18), [noma-security-19](#fact-noma-security-19) | | Stops the action | 6 of 10 | 20 | Block, mask, steer and killswitch actions are described, but available enforcement depends on the integration. | [noma-security-20](#fact-noma-security-20), [noma-security-21](#fact-noma-security-21), [noma-security-23](#fact-noma-security-23), [noma-security-41](#fact-noma-security-41) | | Visibility | 7 of 10 | 15 | MCP server and tool inventory, blast-radius mapping and full interaction observability are described; vendor claims only. | [noma-security-25](#fact-noma-security-25), [noma-security-28](#fact-noma-security-28), [noma-security-30](#fact-noma-security-30), [noma-security-31](#fact-noma-security-31) | | Deployment | 5 of 10 | 10 | API, EDR or MDM discovery and SaaS or on-prem options are described; OS list and time to deploy none found. | [noma-security-08](#fact-noma-security-08), [noma-security-13](#fact-noma-security-13), [noma-security-33](#fact-noma-security-33), [noma-security-34](#fact-noma-security-34), [noma-security-29](#fact-noma-security-29) | | Policy and governance | 5 of 10 | 10 | IdP-scoped policy, framework mapping and audit reporting exist; named SIEM, SOAR, ticketing or EDR export none found. | [noma-security-17](#fact-noma-security-17), [noma-security-32](#fact-noma-security-32), [noma-security-37](#fact-noma-security-37), [noma-security-48](#fact-noma-security-48), [noma-security-35](#fact-noma-security-35) | | Transparency | 3 of 10 | 10 | Vulnerability research is published; a price and public documentation none found. | [noma-security-42](#fact-noma-security-42), [noma-security-43](#fact-noma-security-43), [noma-security-44](#fact-noma-security-44), [noma-security-45](#fact-noma-security-45) | | Proof | 7 of 10 | 5 | Third parties report $132 million raised; named customers none found and the Gartner listing is vendor-cited. | [noma-security-03](#fact-noma-security-03), [noma-security-54](#fact-noma-security-54), [noma-security-05](#fact-noma-security-05), [noma-security-06](#fact-noma-security-06) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 13th | 54.6 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 12th | 57.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 16th | 42.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 9th | 60.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 9th | 57.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 13th | 57.0 | ## Product, deployment, price and owner - Product assessed: Noma Agent Security and Governance Platform: AI-SPM, Agent Access Control, AI-DR for endpoint, SaaS and cloud agents - Deployment: SaaS or on-prem; discovery via APIs, existing EDR or MDM; enforcement via hooks, gateways, SDKs and APIs - Time to deploy: none found on 2026-11-16 - Price: none found on 2026-11-16 - Owner: Noma Security (standalone company; no parent stated on its pages) - Website: [noma.security](https://noma.security) ## Facts and sources - **noma-security-01** (vendor): Noma Security describes its offering as an agent security and governance platform for enterprise security teams.Quote: "The Agent Security & Governance Platform that helps enterprise security teams enable agent adoption without losing control."Source: [noma.security](https://noma.security/), read 2026-11-16. - **noma-security-02** (vendor): Noma says it is an AI security platform built for enterprise AI adoption; its about page lists Niv Braun as Co-Founder and CEO and Alon Tron as Co-Founder and CTO.Quote: "Noma is the AI security platform purpose-built to secure enterprise AI adoption."Source: [noma.security](https://noma.security/about), read 2026-11-16. - **noma-security-03** (third-party): TechCrunch reported on 2024-10-31 that Noma, founded in 2023, closed a $25 million Series A led by Ballistic Ventures, for $32 million raised in total, after a $7 million seed.Quote: "Noma raised $7 million in a previously undisclosed seed round led by Glilot Capital Partners with participation from Cyber Club London."Source: [techcrunch.com](https://techcrunch.com/2024/10/31/noma-is-building-tools-to-spot-security-issues-with-ai-apps/), read 2026-11-16. - **noma-security-04** (vendor): In a press release dated 2026-10-28, Noma lists its backers as Evolution Equity Partners, Ballistic Ventures, Glilot Capital, Cyber Club London, Databricks Ventures and SVCI.Quote: "Backed by Evolution Equity Partners, Ballistic Ventures, Glilot Capital, Cyber Club London, Databricks Ventures, and SVCI"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-named-to-fourth-annual-fortune-cyber-60-list-of-venture-backed-cybersecurity-companies-302922012.html), read 2026-11-16. - **noma-security-05** (vendor): Noma says on its homepage that over 10% of the Fortune 100 trusts it; no individual customer names were shown on the pages read.Quote: "Trusted by over 10% of the Fortune 100"Source: [noma.security](https://noma.security/), read 2026-11-16. - **noma-security-06** (vendor): Noma says Gartner named it a Market Shaper in the 2026 Emerging Market Quadrant for AI Application Security, Startup Vendors.Quote: "Noma named a Market Shaper in the 2026 Gartner® Emerging Market Quadrant for AI Application Security, Startup Vendors"Source: [noma.security](https://noma.security/gartner/2026-gartner-market-quadrant), read 2026-11-16. - **noma-security-07** (vendor): On 2026-10-28 Noma announced it was named to the 2026-27 Fortune Cyber 60 list (early growth category), presented by Lightspeed Venture Partners and Amazon Web Services.Quote: "today announced it has been named to the 2026-27 Fortune Cyber 60, a list presented by Lightspeed Venture Partners and Amazon Web Services"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-named-to-fourth-annual-fortune-cyber-60-list-of-venture-backed-cybersecurity-companies-302922012.html), read 2026-11-16. - **noma-security-08** (vendor): Noma says it discovers endpoint agents, MCP servers and skills through the customer existing EDR or MDM, with no extra endpoint agent.Quote: "Noma discovers agents, MCP servers, and skills through your existing EDR or MDM. See where personal accounts and unapproved tools connect to company resources, without deploying another endpoint agent."Source: [noma.security](https://noma.security/platform/endpoint-agents), read 2026-11-16. - **noma-security-09** (vendor): On 2026-09-28 Noma announced endpoint agent security, saying enforcement coverage spans Claude Code, Claude Cowork, Cursor, Codex, Windsurf, Kiro, Antigravity and OpenClaw.Quote: "with coverage spanning agents including Claude Code, Claude Cowork, Cursor, Codex, Windsurf, Kiro, Antigravity, and OpenClaw"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html), read 2026-11-16. - **noma-security-10** (none-found): None found on 2026-11-16: browser agents, AI browsers or AI browser extensions as a named coverage area on the endpoint agents page.Source: [noma.security](https://noma.security/platform/endpoint-agents), read 2026-11-16. - **noma-security-11** (vendor): Noma's 2026-09-28 release mentions Claude Cowork as an agent that works with browsers, but names no browser-layer control.Quote: "Business users rely on productivity agents such as Claude Cowork to work with documents, browsers, SaaS applications, and internal data."Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html), read 2026-11-16. - **noma-security-12** (vendor): Noma says it discovers agents on SaaS platforms including Agentforce, Microsoft Copilot and Claude Enterprise, and prevents dangerous runtime actions.Quote: "Discover agents across Agentforce, Microsoft Copilot, Claude Enterprise and other SaaS platforms."Source: [noma.security](https://noma.security/platform/saas-agents), read 2026-11-16. - **noma-security-13** (vendor): Noma says it connects through APIs to cloud providers, data platforms, model registries, version control and notebooks to inventory cloud agents.Quote: "Noma connects through APIs to cloud providers, data platforms, model registries, version control, and notebooks"Source: [noma.security](https://noma.security/platform/homegrown-agents), read 2026-11-16. - **noma-security-14** (vendor): Noma names Microsoft Copilot Studio, Salesforce Agentforce and ServiceNow as SaaS agent platforms it covers, plus AWS Bedrock, Azure AI Foundry and Databricks for cloud agents.Quote: "Business analysts, sales ops, HR, and finance teams build agents on Microsoft Copilot Studio, Salesforce Agentforce, and ServiceNow, often without telling security."Source: [noma.security](https://noma.security/platform), read 2026-11-16. - **noma-security-15** (vendor): Noma says it maps each SaaS agent to the human who built it and the permissions it inherited.Quote: "Noma maps every agent to the human who built it and the permissions it inherited, and displays both in the agent risk map."Source: [noma.security](https://noma.security/platform/saas-agents), read 2026-11-16. - **noma-security-16** (vendor): In a release dated 2026-06-02 Noma said Agent Access Control gives each agent a distinct, attributable identity when it connects to MCP servers and tools.Quote: "Noma Agent Access Control gives each autonomous agent a distinct, attributable identity when it connects to MCP servers and tools."Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-launches-agentic-access-control-to-govern-ai-agents-and-mcp-servers-across-the-enterprise-302788534.html), read 2026-11-16. - **noma-security-17** (vendor): On 2026-09-28 Noma said it ties each agent to a human and to identity provider users and groups so policy can apply by user, group, tool or organization.Quote: "Noma attributes each agent to the human behind it and connects that identity to IdP users and groups"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html), read 2026-11-16. - **noma-security-18** (vendor): Noma says it enforces policy through AI gateways, MCP gateways, agent hooks, agent SDKs and direct APIs already in the customer environment.Quote: "Decouple governance from any single control point and enforce policy through AI gateways, MCP gateways, agent hooks, agent SDKs, and direct APIs, using the infrastructure already in your environment."Source: [noma.security](https://noma.security/solutions/mcp-security), read 2026-11-16. - **noma-security-19** (vendor): Noma says one policy set is enforced for agents on Bedrock behind an AI gateway, on Copilot Studio via integrations, and on developer machines via native hooks in Claude Code or Cursor.Quote: "The same behavioral monitoring and policy enforcement applies whether the agent runs on Bedrock behind an AI gateway"Source: [noma.security](https://noma.security/solutions/agent-behavior-runtime-control), read 2026-11-16. - **noma-security-20** (vendor): Noma says AI-DR policies can block an action, mask sensitive data, steer the agent, bring in a human or trigger a killswitch.Quote: "block an action, mask sensitive data, steer the agent towards a different path, get a human in the loop, or trigger a killswitch."Source: [noma.security](https://noma.security/products/ai-dr), read 2026-11-16. - **noma-security-21** (vendor): Noma says its response policy can alert, block or mask data, and that available enforcement actions depend on the integration.Quote: "Available enforcement actions depend on the integration."Source: [noma.security](https://noma.security/solutions/agent-behavior-runtime-control), read 2026-11-16. - **noma-security-22** (vendor): On 2026-09-28 Noma said AI-DR detects prompt injection, sensitive data leakage, malicious intent, tool poisoning, scope violations and drift from intent.Quote: "detects prompt injection, sensitive data leakage, malicious intent, tool poisoning, scope violations, and agents drifting from their intent or established behavior"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html), read 2026-11-16. - **noma-security-23** (vendor): Noma says it restricts individual tools in approved MCP servers and enforces the decision when agents attempt to connect or act.Quote: "Restrict individual tools within approved servers and enforce those decisions when agents attempt to connect or act."Source: [noma.security](https://noma.security/solutions/mcp-security), read 2026-11-16. - **noma-security-25** (vendor): Noma says it shows which MCP servers are installed, which agents use them and which tools they expose, including unapproved servers.Quote: "See which MCP servers are installed, which agents use them, and which tools they expose."Source: [noma.security](https://noma.security/solutions/mcp-security), read 2026-11-16. - **noma-security-28** (vendor): Noma says AI-SPM inspects MCP servers, models and skills for risk and keeps a live registry of allowed AI components.Quote: "Inspect MCP servers, models, and skills for risk in the agent supply chain, and build a live registry of which AI components are allowed in the organization."Source: [noma.security](https://noma.security/products/ai-spm), read 2026-11-16. - **noma-security-29** (none-found): None found on 2026-11-16: supported operating systems or a time to deploy, on the endpoint agents and MCP security pages.Source: [noma.security](https://noma.security/platform/endpoint-agents), read 2026-11-16. - **noma-security-30** (vendor): Noma says it maps each agent blast radius across skills, tools, MCPs, models, data sources, channels and other agents.Quote: "Map the full blast radius of each agent, including skills, tools, MCPs, models, data sources, communication channels, and other agents."Source: [noma.security](https://noma.security/solutions/ai-asset-discovery), read 2026-11-16. - **noma-security-31** (vendor): Noma says AI-DR gives observability into every prompt, tool call, response and interaction across endpoint, SaaS and cloud agents.Quote: "complete observability into every prompt, tool call, response, and interaction your agents engage in across endpoint, SaaS, and cloud environments"Source: [noma.security](https://noma.security/products/ai-dr), read 2026-11-16. - **noma-security-32** (vendor): Noma says AI-SPM maps findings to NIST AI RMF, EU AI Act, ISO 42001, OWASP LLM Top 10 and MITRE ATLAS.Quote: "Map findings to NIST AI RMF, EU AI Act, ISO 42001, OWASP LLM Top 10, and MITRE ATLAS."Source: [noma.security](https://noma.security/products/ai-spm), read 2026-11-16. - **noma-security-33** (vendor): Noma says it discovers SaaS agents through platform APIs with nothing to deploy.Quote: "Noma connects to SaaS agent platforms through their APIs and discovers every agent continuously, with nothing to deploy."Source: [noma.security](https://noma.security/platform/saas-agents), read 2026-11-16. - **noma-security-34** (vendor): Noma says it supports both on-prem and SaaS deployments.Quote: "Support for both on-prem and SaaS deployments, so no model, training data, or security events leave your environment."Source: [noma.security](https://noma.security/solutions/mcp-security), read 2026-11-16. - **noma-security-35** (none-found): None found on 2026-11-16: named SIEM, SOAR, ticketing or EDR export integrations (beyond EDR/MDM discovery) on the AI-DR page.Source: [noma.security](https://noma.security/products/ai-dr), read 2026-11-16. - **noma-security-36** (vendor): Noma says it has integrations with 80+ services; the services are not listed on the page read.Quote: "With integrations with 80+ services and coverage spanning endpoint, SaaS, and cloud agents"Source: [noma.security](https://noma.security/solutions/mcp-security), read 2026-11-16. - **noma-security-37** (vendor): Noma says policy can be scoped by server, individual tool and user group.Quote: "Noma scopes policy by server, by individual tool, and by user group"Source: [noma.security](https://noma.security/solutions/mcp-security), read 2026-11-16. - **noma-security-41** (vendor): Noma says AI-DR evaluates each cloud agent action and, based on policy, alerts, blocks, masks data or routes to a human.Quote: "Based on policy, it alerts, blocks, masks data, or routes to a human."Source: [noma.security](https://noma.security/platform/homegrown-agents), read 2026-11-16. - **noma-security-42** (none-found): None found on 2026-11-16: a published price; noma.security/pricing returns Not Found and the sitemap lists no pricing page.Source: [noma.security](https://noma.security/pricing), read 2026-11-16. - **noma-security-43** (none-found): None found on 2026-11-16: public documentation; docs.noma.security redirects to a sign-in page.Source: [docs.noma.security](https://docs.noma.security), read 2026-11-16. - **noma-security-44** (vendor): Noma Labs published research on 2026-10-14 on Claude Code Desktop; the vendor says Anthropic scored the finding CVSS 8.7.Quote: "Anthropic scored Noma Labs' discovery as CVSS 8.7 (High) and treated the impact as equivalent to code execution on the host."Source: [noma.security](https://noma.security/noma-labs/sneakpeek-how-the-preview-pane-in-claude-code-desktop-can-exfiltrate-your-data), read 2026-11-16. - **noma-security-45** (vendor): Noma announced RufRoot, a CVSS 10.00 flaw its Labs found in the Ruflo agent platform (CVE-2026-59726); the release text read gives a 2026-06-30 publication date.Quote: "a maximum severity (CVSS 10.00) vulnerability discovered by Noma Labs in Ruflo"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/noma-labs-discovers-critical-vulnerability-in-widely-adopted-open-source-ai-agent-platform-ruflo-302837494.html), read 2026-11-16. - **noma-security-47** (vendor): Noma says AI-SPM finds misconfigurations such as excessive agency, unsandboxed agents, secrets in instructions and unauthenticated access.Quote: "Find risky mistakes and misconfigurations in AI agents, like excessive agency, unsandboxed agents, secrets in agent instructions, and unauthenticated access."Source: [noma.security](https://noma.security/products/ai-spm), read 2026-11-16. - **noma-security-48** (vendor): Noma says reporting includes inventory, framework mappings, test results and audit trails.Quote: "inventory, framework mappings, test results, and audit trails."Source: [noma.security](https://noma.security/products/ai-dr), read 2026-11-16. - **noma-security-51** (vendor): Noma says it flags the combination of untrusted input, sensitive data access and an external send as a session-level risk.Quote: "Untrusted input, sensitive data access, and an external send can each be fine alone and together form the lethal trifecta."Source: [noma.security](https://noma.security/solutions/agent-behavior-runtime-control), read 2026-11-16. - **noma-security-54** (third-party): SiliconANGLE reported on 2025-07-31 that Noma raised a $100 million Series B led by Evolution Equity Partners with Ballistic Ventures and Glilot Capital.Quote: "Evolution Equity Partners led the Series B investment with participation from Ballistic Ventures and Glilot Capital."Source: [siliconangle.com](https://siliconangle.com/2025/07/31/ai-security-startup-noma-security-closes-100m-funding-round/), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Microsoft profile](/vendors/microsoft/) - [Operant AI profile](/vendors/operant-ai/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Operant AI agent security URL: https://aiagentsecuritysolutions.com/vendors/operant-ai/ Updated: 2026-11-16 Operant AI is the only vendor here with documented coverage at a working level on all five layers. Per its pages, Endpoint Protector covers AI IDEs, coding agents, desktop and MCP clients through MDM-rolled clients on macOS, Windows and Linux, and blocks out-of-scope tool calls on the device. Browser Protection, announced 27 Aug 2026, puts allow, sanitize or block inline. Agent Protector inventories cloud and SaaS agents, and an MCP gateway builds trust zones that block untrusted servers. Operant also names exfiltration, credential access, package installs and payments as blocked actions. Public documentation and a published price: none found on 16 Nov 2026, and the Microsoft Copilot path detects but does not block inline. ## Strength and weakness for AI Agent Security Solutions **Strength:** Documents agent coverage on endpoints, browsers, Microsoft 365 and cloud agents, and an MCP gateway, with inline blocking (operant-ai-09, operant-ai-13, operant-ai-15, operant-ai-21, operant-ai-23). **Weakness:** No public documentation site, no published price, and no named paying customers list: none found on 2026-11-16 (operant-ai-41, operant-ai-42, operant-ai-08). ## Scores and evidence Operant AI is 1st in the overall ranking at 64.9 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Operant AI: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 8 of 10 | 6 | Vendor documentation covers coding agents, desktop and MCP clients with MDM-rolled clients and on-device blocking; a third-party report confirms the endpoint launch. | [operant-ai-09](#fact-operant-ai-09), [operant-ai-10](#fact-operant-ai-10), [operant-ai-11](#fact-operant-ai-11), [operant-ai-45](#fact-operant-ai-45) | | Coverage: browser layer | 6 of 10 | 6 | The vendor documents an AI browser protection product with inline allow, sanitize or block; detail on browser agents and supported browsers is thin. | [operant-ai-12](#fact-operant-ai-12), [operant-ai-13](#fact-operant-ai-13) | | Coverage: cloud workspace layer | 8 of 10 | 6 | The vendor inventories cloud and SaaS agents and the observed integrations directory lists AgentCore, Copilot Studio and Agentforce; the Copilot path detects only and is not inline. | [operant-ai-14](#fact-operant-ai-14), [operant-ai-15](#fact-operant-ai-15), [operant-ai-16](#fact-operant-ai-16), [operant-ai-17](#fact-operant-ai-17) | | Coverage: identity layer | 5 of 10 | 6 | The vendor catalogs agent identities and lists access enforcement; detail on least privilege review and revocation is none found. | [operant-ai-18](#fact-operant-ai-18), [operant-ai-19](#fact-operant-ai-19), [operant-ai-20](#fact-operant-ai-20) | | Coverage: gateway layer | 7 of 10 | 6 | The vendor documents an LLM gateway with a firewall plus MCP trust zones that block untrusted servers and tools; no second source found. | [operant-ai-21](#fact-operant-ai-21), [operant-ai-22](#fact-operant-ai-22) | | Stops the action | 7 of 10 | 20 | The vendor names blocking of exfiltration, credential access, package installs, out-of-scope tool calls and payments before execution; vendor claims only. | [operant-ai-23](#fact-operant-ai-23), [operant-ai-24](#fact-operant-ai-24), [operant-ai-27](#fact-operant-ai-27), [operant-ai-30](#fact-operant-ai-30), [operant-ai-31](#fact-operant-ai-31) | | Visibility | 7 of 10 | 15 | The vendor describes a live inventory mapped to user and device with traces from prompts through tool calls; vendor claims only. | [operant-ai-09](#fact-operant-ai-09), [operant-ai-14](#fact-operant-ai-14), [operant-ai-39](#fact-operant-ai-39), [operant-ai-40](#fact-operant-ai-40) | | Deployment | 7 of 10 | 10 | The vendor publishes MDM rollout on three OSes, a helm install and times under 15 and 5 minutes; vendor claims only. | [operant-ai-10](#fact-operant-ai-10), [operant-ai-34](#fact-operant-ai-34), [operant-ai-35](#fact-operant-ai-35) | | Policy and governance | 7 of 10 | 10 | Natural-language policy with audit reasons, SIEM forwarding and Okta, Splunk and CrowdStrike integrations are listed; ticketing integration none found. | [operant-ai-11](#fact-operant-ai-11), [operant-ai-17](#fact-operant-ai-17), [operant-ai-36](#fact-operant-ai-36), [operant-ai-37](#fact-operant-ai-37), [operant-ai-38](#fact-operant-ai-38) | | Transparency | 3 of 10 | 10 | Product pages, open-source code and research are public; public documentation and a published price are none found. | [operant-ai-41](#fact-operant-ai-41), [operant-ai-42](#fact-operant-ai-42), [operant-ai-43](#fact-operant-ai-43), [operant-ai-44](#fact-operant-ai-44) | | Proof | 6 of 10 | 5 | The vendor shows modest funding, testimonials, a CRN listing and a marketplace profile; named paying customers and independent analyst reports are none found. | [operant-ai-03](#fact-operant-ai-03), [operant-ai-05](#fact-operant-ai-05), [operant-ai-07](#fact-operant-ai-07), [operant-ai-08](#fact-operant-ai-08) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 1st | 64.9 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 3rd | 68.5 | | [Ranked for the browser layer](/best-for/layer-browser/) | joint 3rd | 62.5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | joint 1st | 68.5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 8th | 59.5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 2nd | 65.5 | ## Product, deployment, price and owner - Product assessed: Agent Protector, Endpoint Protector, MCP Gateway, AI Gatekeeper and Browser Protection - Deployment: Endpoint clients rolled out by MDM; Kubernetes helm install for cloud and API protection; gateway for LLM and MCP traffic; VPC, on-prem and air-gapped options on Enterprise - Time to deploy: Vendor says endpoint rollout in under 15 minutes and helm install in under 5 minutes - Price: No prices published; three plans (Pro, Scale, Enterprise) with a 7-day trial; none found on 2026-11-16 for amounts - Owner: Operant AI, Inc. (standalone company, San Francisco) - Website: [operant.ai](https://www.operant.ai) ## Facts and sources - **operant-ai-01** (vendor): Operant AI describes itself as a runtime defense platform for live cloud and AI applications; its footer names Operant AI, Inc. and San Francisco.Quote: "Operant AI is the only Runtime AI Application Defense Platform that actively protects every layer of live cloud and AI applications"Source: [operant.ai](https://www.operant.ai/company/about), read 2026-11-16. - **operant-ai-02** (vendor): The homepage footer shows the owner as Operant AI, Inc., a standalone company, with SOC 2 Type II compliance stated.Quote: "Operant is SOC 2, Type II Compliant."Source: [operant.ai](https://www.operant.ai), read 2026-11-16. - **operant-ai-03** (vendor): Operant says its Series A of $10 million, led by SineWave Ventures and Felicis, brought total funding to $13.5 million (post dated 2024).Quote: "Operant's $10 million Series A brings Operant's total funding to $13.5 million."Source: [operant.ai](https://www.operant.ai/art-kubed/operant-ai-series-a), read 2026-11-16. - **operant-ai-04** (vendor): Operant's about page lists Vrajesh Bhavsar as Co-Founder and CEO and Priyanka Tembey as Co-Founder and CTO.Quote: "Co-Founder and CEO"Source: [operant.ai](https://www.operant.ai/company/about), read 2026-11-16. - **operant-ai-05** (third-party): CRN's 2026-07-02 list of the 10 hottest AI security startups includes Operant AI, founded 2021, CEO Vrajesh Bhavsar.Quote: "Operant AI Founded: 2021 CEO: Vrajesh Bhavsar"Source: [crn.com](https://www.crn.com/news/security/2026/the-10-hottest-ai-security-startups-of-2026-so-far?page=5), read 2026-11-16. - **operant-ai-06** (vendor): Operant says it is the only vendor featured across five Gartner AI security reports (vendor claim, not on its own checked).Quote: "Operant AI is the only vendor featured across all five of Gartner®'s most critical AI security reports"Source: [operant.ai](https://www.operant.ai), read 2026-11-16. - **operant-ai-07** (observed): Operant's AWS Marketplace seller profile exists and repeats its runtime defense positioning; the page lists Operant as an AWS Marketplace seller.Quote: "Operant AI is the world's only Runtime Application Defense Platform."Source: [aws.amazon.com](https://aws.amazon.com/marketplace/seller-profile?id=seller-gebxlkhuqojto), read 2026-11-16. - **operant-ai-08** (vendor): The homepage shows attributed testimonials (for example Chargebee, Juniper Networks, Cohere) but no list of named paying customers; none found on 2026-11-16 beyond testimonial attributions.Quote: "Cyber Security Leader | Chargebee"Source: [operant.ai](https://www.operant.ai), read 2026-11-16. - **operant-ai-09** (vendor): Operant says Endpoint Protector inspects AI IDEs, coding agents, desktop clients, MCP clients, skills, tools and plugins on employee endpoints (generally available).Quote: "Endpoint Protector inspects every AI IDE, coding agent, desktop client, MCP client, skill, tool, and plugin"Source: [operant.ai](https://www.operant.ai/platform/endpoint-protector), read 2026-11-16. - **operant-ai-10** (vendor): Operant says Endpoint Protector ships native clients for macOS, Windows and Linux, rolled out by MDM (Jamf, Intune, Workspace ONE, Kandji), with workforce rollout in under 15 minutes.Quote: "Workforce-wide rollout in under 15 minutes."Source: [operant.ai](https://www.operant.ai/platform/endpoint-protector), read 2026-11-16. - **operant-ai-11** (vendor): Operant says Endpoint Protector enforces role-based policy with IdP inheritance and ScopeGuard blocking out-of-scope tool calls on the device.Quote: "ScopeGuard blocks out-of-scope tool calls in real time"Source: [operant.ai](https://www.operant.ai/platform/endpoint-protector), read 2026-11-16. - **operant-ai-12** (vendor): Operant has a browser protection page for AI browsers and gen AI websites, enforced by a local engine on the device.Quote: "Your browser has access to everything you're logged into. Now it takes orders from the pages it reads."Source: [operant.ai](https://www.operant.ai/solutions/browser-protection), read 2026-11-16. - **operant-ai-13** (vendor): On 2026-08-27, Operant announced Browser Protection that puts an allow, sanitize or block decision inline on the browser, in both directions, with the same policy engine and audit trail.Quote: "Operant Browser Protection puts the same allow/sanitize/block decision inline on that surface, in both directions"Source: [operant.ai](https://www.operant.ai/art-kubed/introducing-operant-semantic-firewall), read 2026-11-16. - **operant-ai-14** (vendor): Operant says Agent Protector inventories managed and unmanaged agents in cloud environments, SaaS platforms and development tools.Quote: "This includes managed agents deployed through official channels, as well as unmanaged agents running in cloud environments, SaaS platforms, and development tools."Source: [operant.ai](https://www.operant.ai/platform/agent-protector), read 2026-11-16. - **operant-ai-15** (vendor): On 2026-10-01, Operant announced inline defense for custom agents built on the Microsoft 365 Agents SDK plus Purview-based observability for Copilot; it states the Purview path detects but does not sit inline.Quote: "Agent Protector goes inside the agent loop of custom Copilot agents, so each step is evaluated before it executes."Source: [operant.ai](https://www.operant.ai/art-kubed/announcing-operants-coverage-for-microsoft-copilot-and-every-ai-assistant-embedded-in-microsoft-365), read 2026-11-16. - **operant-ai-16** (vendor): Operant's pricing page says Agent Protector gives runtime security for homegrown or cloud-native cloud agents with support for AWS, Azure and more.Quote: "Runtime security for cloud agents, whether homegrown or cloud-native. In-depth support for AWS, Azure, and more."Source: [operant.ai](https://www.operant.ai/platform/pricing), read 2026-11-16. - **operant-ai-17** (observed): Operant's integrations directory lists agent platforms and tools including AWS AgentCore, Microsoft Copilot Studio (shown), Salesforce AgentForce, Okta, Splunk, CrowdStrike and SentinelOne.Quote: "Microsoft Copilot Studio Agent Frameworks"Source: [operant.ai](https://www.operant.ai/platform/integrations), read 2026-11-16. - **operant-ai-18** (vendor): Operant says Agent Protector catalogs agent identities including user and service accounts.Quote: "The system creates detailed catalogs of agent identities, including both user and service accounts"Source: [operant.ai](https://www.operant.ai/platform/agent-protector), read 2026-11-16. - **operant-ai-19** (vendor): Operant's pricing page lists agent identity and access enforcement and SIEM, SOC/SOAR integrations in the Scale plan.Quote: "Agent identity & access enforcement"Source: [operant.ai](https://www.operant.ai/platform/pricing), read 2026-11-16. - **operant-ai-20** (vendor): Operant says MCP Gateway supports MCP and AI non-human identities with detection and access control.Quote: "get comprehensive support for Model Context Protocol (MCP), and AI Non-Human Identities (NHIs)"Source: [operant.ai](https://www.operant.ai/platform/ai-gatekeeper), read 2026-11-16. - **operant-ai-21** (vendor): Operant says MCP Gateway builds MCP trust zones, blocks untrusted servers and tools in real time, and enforces least privilege on tool use.Quote: "Implement MCP trust zones with real-time blocking of untrusted servers and tools."Source: [operant.ai](https://www.operant.ai/solutions/mcp-gateway), read 2026-11-16. - **operant-ai-22** (vendor): Operant says AI Gatekeeper is an LLM gateway with a native firewall, multi-provider routing and request inspection for AI applications.The exact quote is on the source page.Source: [operant.ai](https://www.operant.ai/platform/pricing), read 2026-11-16. - **operant-ai-23** (vendor): Operant says Agent Protector blocks inline when an agent goes outside established patterns, escalates privileges or shows data exfiltration.Quote: "Agent Protector blocks the action inline."Source: [operant.ai](https://www.operant.ai/platform/agent-protector), read 2026-11-16. - **operant-ai-24** (vendor): Operant says Semantic Firewall's Tool Intent Guard blocks exfiltration, bulk dumps, credential access and unauthorized sharing before execution.Quote: "blocks exfiltration, bulk dumps, credential access, and unauthorized sharing"Source: [operant.ai](https://www.operant.ai/platform/semantic-firewall), read 2026-11-16. - **operant-ai-25** (vendor): On 2026-08-27, Operant launched Semantic Firewall, which returns allow, block or redact decisions as the agent acts, with a plain-language reason for each; no numeric latency is given on that page.Quote: "returns an allow, block, or redact decision as the agent acts"Source: [operant.ai](https://www.operant.ai/art-kubed/introducing-operant-semantic-firewall), read 2026-11-16. - **operant-ai-27** (vendor): Operant says Endpoint Protector's CodeInjectionGuard blocks malicious payloads, prompt-injected commands and unsanctioned package installs before they run.Quote: "blocking malicious payloads, prompt-injected commands, and unsanctioned package installs before they run"Source: [operant.ai](https://www.operant.ai/platform/endpoint-protector), read 2026-11-16. - **operant-ai-30** (vendor): Operant says Agent ScopeGuard blocks out-of-scope actions before execution, limiting the data sources and APIs an agent may use.Quote: "blocking out-of-scope actions before execution and alerting your security team with full action traces"Source: [operant.ai](https://www.operant.ai/platform/agent-protector), read 2026-11-16. - **operant-ai-31** (vendor): On 2026-10-15, Operant said Agent Protector's Agentic Commerce Defense evaluates each agent payment attempt and can block a payment before it is submitted.Quote: "Operant can block the payment before it is submitted."Source: [operant.ai](https://www.operant.ai/art-kubed/stop-dollar-exfiltration-operant-agent-protector-now-defends-agentic-commerce), read 2026-11-16. - **operant-ai-34** (vendor): Operant says its platform deploys with a single-step helm install on Kubernetes 1.25+, works in under 5 minutes, and needs zero instrumentation.Quote: "Single step helm install. Zero instrumentation. Zero Integrations. Works in <5 minutes."Source: [operant.ai](https://www.operant.ai/platform/faqs), read 2026-11-16. - **operant-ai-35** (vendor): Operant's Enterprise plan lists VPC, on-prem and air-gapped options plus SSO, SCIM, MDM templates and API/webhooks.Quote: "VPC, on-prem, and air-gapped options"Source: [operant.ai](https://www.operant.ai/platform/pricing), read 2026-11-16. - **operant-ai-36** (vendor): Operant says Endpoint Protector forwards agent-loop telemetry to Splunk, Datadog, Elastic, Chronicle and more.Quote: "Forwards rich telemetry to Splunk, Datadog, Elastic, Chronicle, and more"Source: [operant.ai](https://www.operant.ai/platform/endpoint-protector), read 2026-11-16. - **operant-ai-37** (vendor): Operant says Semantic Firewall policies are written in natural language and every decision carries a plain-language reason for audit.Quote: "Every decision carries a plain-language reason that stands up to review."Source: [operant.ai](https://www.operant.ai/platform/semantic-firewall), read 2026-11-16. - **operant-ai-38** (vendor): Operant's pricing lists standard audit logs in Pro and SIEM, SOC/SOAR integrations and extended retention from the Scale plan.Quote: "SIEM, SOC/SOAR integrations"Source: [operant.ai](https://www.operant.ai/platform/pricing), read 2026-11-16. - **operant-ai-39** (vendor): Operant says Endpoint Protector builds a live inventory of AI tools, models, MCP servers and skills and maps each to user, device and risk score.Quote: "Maps each tool to user, device, business unit, and risk score"Source: [operant.ai](https://www.operant.ai/platform/endpoint-protector), read 2026-11-16. - **operant-ai-40** (vendor): Operant says Agent Protector traces agents from prompts through tool calls to memory access, with activity timelines and tool activity graphs.Quote: "comprehensive tracing from initial prompts through tool calls to memory store access"Source: [operant.ai](https://www.operant.ai/platform/agent-protector), read 2026-11-16. - **operant-ai-41** (vendor): The product pages are readable without sign-in, but the pricing page lists plans without prices and says to talk to the vendor.Quote: "All plans support multi-year commitments with volume discounts. Talk to us for details."Source: [operant.ai](https://www.operant.ai/platform/pricing), read 2026-11-16. - **operant-ai-42** (none-found): None found on 2026-11-16: public product documentation. docs.operant.ai redirects to the product sign-in page.Source: [docs.operant.ai](https://docs.operant.ai), read 2026-11-16. - **operant-ai-43** (observed): Operant publishes Woodpecker, an open-source red teaming tool, on GitHub under the Apache-2.0 license.Quote: "woodpecker is a modular red teaming tool focused for AI and cloud apps."Source: [github.com](https://github.com/OperantAI/woodpecker), read 2026-11-16. - **operant-ai-44** (third-party): GBHackers (October 2025) reports Operant's research team found the Shadow Escape zero-click MCP attack and that Operant reported it to OpenAI and filed a CVE.Quote: "Operant AI has reported this attack to OpenAI and filed a CVE to address this emerging threat to data governance and privacy."Source: [gbhackers.com](https://gbhackers.com/zero-click-exploit-targets-ai-agents/), read 2026-11-16. - **operant-ai-45** (third-party): CyberTech Intelligence reported on the Endpoint Protector launch (Operant's news page dates it 2026-05-04) as a way to secure AI interactions at the endpoint.Quote: "Operant AI has launched Endpoint Protector, a new solution designed to secure AI interactions directly at the endpoint."Source: [cybertechnologyinsights.com](https://cybertechnologyinsights.com/endpoint-security/operant-ai-introduces-endpoint-security-for-ai-tools/), read 2026-11-16. - **operant-ai-46** (vendor): Operant's news page shows 2026-07-02 CRN coverage and Gartner MCP guide coverage dated 2026-01-02.Quote: "Gartner features Operant AI's MCP Gateway in its newest MCP cybersecurity guide"Source: [operant.ai](https://www.operant.ai/company/news), read 2026-11-16. ## News mentioning Operant AI - [Operant extends agent defense to Microsoft 365 Copilot](/news/operant-ai-microsoft-365-copilot-agents-sdk-coverage/) ## Related pages - [All vendor profiles](/vendors/) - [Noma Security profile](/vendors/noma-security/) - [Palo Alto Networks profile](/vendors/palo-alto-networks/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Palo Alto Networks AI agent security URL: https://aiagentsecuritysolutions.com/vendors/palo-alto-networks/ Updated: 2026-11-16 Palo Alto Networks documents SaaS Agent Security for agents built on platforms such as Microsoft Copilot Studio and ServiceNow, with remediation, and a Prisma AIRS AI Gateway that sits between users or applications and each LLM provider with inline guardrails. An MCP scanning server is public in a code repository, which confirms the claim. Prisma Browser and a governance layer for agentic browsing cover the browser, and an October 2026 update brings unified identity to agents. Endpoint coverage is limited to best-effort example hooks for coding assistants, and SIEM export: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers SaaS-built agents, agent identity, a gateway and browsers under one platform with block actions (palo-alto-networks-33, palo-alto-networks-38, palo-alto-networks-41, palo-alto-networks-47). **Weakness:** Endpoint coverage is limited to best-effort example hooks, and none found on 2026-11-16 for SIEM export (palo-alto-networks-50, palo-alto-networks-67). ## Scores and evidence Palo Alto Networks is 5th in the overall ranking at 61.4 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Palo Alto Networks: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 3 of 10 | 6 | Only best-effort community example hooks for coding assistants are observed; endpoint agent coverage none found in the product. | [palo-alto-networks-04](#fact-palo-alto-networks-04), [palo-alto-networks-50](#fact-palo-alto-networks-50), [palo-alto-networks-51](#fact-palo-alto-networks-51) | | Coverage: browser layer | 6 of 10 | 6 | Extension discovery and blocking and agentic browsing governance are listed; the third-party report is older and not agent-specific. | [palo-alto-networks-08](#fact-palo-alto-networks-08), [palo-alto-networks-09](#fact-palo-alto-networks-09), [palo-alto-networks-49](#fact-palo-alto-networks-49), [palo-alto-networks-70](#fact-palo-alto-networks-70) | | Coverage: cloud workspace layer | 8 of 10 | 6 | SaaS agent security covers Copilot Studio and ServiceNow with remediation, and observed docs confirm cloud agent discovery. | [palo-alto-networks-33](#fact-palo-alto-networks-33), [palo-alto-networks-34](#fact-palo-alto-networks-34), [palo-alto-networks-35](#fact-palo-alto-networks-35) | | Coverage: identity layer | 4 of 10 | 6 | Agent identity governance is described in general terms; credential inventory and revocation detail none found. | [palo-alto-networks-38](#fact-palo-alto-networks-38), [palo-alto-networks-39](#fact-palo-alto-networks-39), [palo-alto-networks-40](#fact-palo-alto-networks-40) | | Coverage: gateway layer | 8 of 10 | 6 | An AI gateway with inline guardrails and an MCP scanning server are documented, with observed repo confirmation. | [palo-alto-networks-41](#fact-palo-alto-networks-41), [palo-alto-networks-42](#fact-palo-alto-networks-42), [palo-alto-networks-45](#fact-palo-alto-networks-45), [palo-alto-networks-51](#fact-palo-alto-networks-51) | | Stops the action | 6 of 10 | 20 | Requests are blocked inline and risky agents paused or terminated; specific agent action types are limited. | [palo-alto-networks-34](#fact-palo-alto-networks-34), [palo-alto-networks-42](#fact-palo-alto-networks-42), [palo-alto-networks-47](#fact-palo-alto-networks-47), [palo-alto-networks-48](#fact-palo-alto-networks-48) | | Visibility | 6 of 10 | 15 | Agent lists with configuration and posture are documented; per-agent tool reach and activity logs are limited. | [palo-alto-networks-33](#fact-palo-alto-networks-33), [palo-alto-networks-35](#fact-palo-alto-networks-35), [palo-alto-networks-36](#fact-palo-alto-networks-36), [palo-alto-networks-37](#fact-palo-alto-networks-37) | | Deployment | 7 of 10 | 10 | Gateway, browser, API and hybrid modes with observed installers are documented; time to deploy for agent products none found. | [palo-alto-networks-14](#fact-palo-alto-networks-14), [palo-alto-networks-16](#fact-palo-alto-networks-16), [palo-alto-networks-17](#fact-palo-alto-networks-17), [palo-alto-networks-44](#fact-palo-alto-networks-44), [palo-alto-networks-52](#fact-palo-alto-networks-52) | | Policy and governance | 5 of 10 | 10 | Per-team policies, logs and ticket creation exist; SIEM, SOAR, IdP and EDR integrations none found. | [palo-alto-networks-43](#fact-palo-alto-networks-43), [palo-alto-networks-46](#fact-palo-alto-networks-46), [palo-alto-networks-66](#fact-palo-alto-networks-66), [palo-alto-networks-67](#fact-palo-alto-networks-67) | | Transparency | 7 of 10 | 10 | Documentation is public (observed) and a research report is published; a published price none found. | [palo-alto-networks-17](#fact-palo-alto-networks-17), [palo-alto-networks-26](#fact-palo-alto-networks-26), [palo-alto-networks-24](#fact-palo-alto-networks-24) | | Proof | 8 of 10 | 5 | A listed company with named customers and third-party coverage; the Gartner listing is vendor-cited. | [palo-alto-networks-28](#fact-palo-alto-networks-28), [palo-alto-networks-29](#fact-palo-alto-networks-29), [palo-alto-networks-30](#fact-palo-alto-networks-30), [palo-alto-networks-32](#fact-palo-alto-networks-32), [palo-alto-networks-70](#fact-palo-alto-networks-70) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 5th | 61.4 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 14th | 53.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 5th | 62.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 3rd | 68.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 10th | 56.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 1st | 68.0 | ## Product, deployment, price and owner - Product assessed: Prisma AIRS (AI Gateway, SaaS Agent Security, Agent Identity Security, AI Runtime Security) with Prisma Browser - Deployment: AI Access Security via existing NGFW and SSE gateways with no endpoint agent; Prisma Browser as a browser, extension or mobile app; Prisma AIRS via API Intercept, Network Intercept and AI Gateway (SaaS or Hybrid) - Time to deploy: none found on 2026-11-16 (Prisma Browser: access 'in minutes' per vendor) - Price: none found on 2026-11-16 - Owner: Palo Alto Networks, Inc. (NASDAQ: PANW) - Website: paloaltonetworks.com ## Facts and sources - **palo-alto-networks-01** (vendor): The Palo Alto Networks leadership page lists Nikesh Arora as chairman and CEO, so the company is its own parent and the owner of Prisma AIRS, AI Access Security and Prisma Browser.Quote: "Nikesh Arora is chairman and CEO of Palo Alto Networks, the world’s leading cybersecurity company."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/about-us/management), read 2026-11-16. - **palo-alto-networks-02** (vendor): On 2026-05-29 Palo Alto Networks said it closed the acquisition of Portkey, an AI gateway company, and made it the AI Gateway of Prisma AIRS.Quote: "today announced it has closed the acquisition of Portkey, a pioneer in AI Gateways."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/company/press/2026/palo-alto-networks-completes-acquisition-of-portkey-to-secure-ai-agents), read 2026-11-16. - **palo-alto-networks-03** (vendor): Palo Alto Networks documents Prisma AIRS as its AI security platform covering AI applications, models, agents and infrastructure across the AI lifecycle.Quote: "Prisma® AIRS™ is Palo Alto Networks' comprehensive AI security platform."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/ai-runtime-security), read 2026-11-16. - **palo-alto-networks-04** (none-found): None found on 2026-11-16: endpoint-agent coverage for desktop or command-line agents inside the Prisma AIRS product page; the page lists Agentic Endpoint Security as a separate capability, which is a separate profile and not counted here.Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/prisma/prisma-ai-runtime-security), read 2026-11-16. - **palo-alto-networks-08** (vendor): Palo Alto Networks says Prisma Browser discovers all extensions in use, monitors them and blocks risky or over-permissioned ones.Quote: "Discover all extensions in use, continuously monitor them for threats, and block risky or over-permissioned extensions."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/sase/prisma-browser), read 2026-11-16. - **palo-alto-networks-09** (vendor): Palo Alto Networks claims Prisma Browser gives visibility into all GenAI apps and agentic workflows.Quote: "Visibility into all GenAI apps and agentic workflows, effectively eliminating all shadow AI."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/sase/prisma-browser), read 2026-11-16. - **palo-alto-networks-14** (vendor): Palo Alto Networks says AI Access Security deploys through existing SSE and firewall gateways with no additional endpoint agents or infrastructure changes.The exact quote is on the source page.Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/sase/ai-access-security), read 2026-11-16. - **palo-alto-networks-16** (vendor): Palo Alto Networks documents Prisma Browser as a browser, an extension for consumer browsers and a mobile app, with access in minutes.Quote: "Available as a browser, an extension for consumer browsers, and a mobile app, it provides secure access from any location in minutes."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-browser), read 2026-11-16. - **palo-alto-networks-17** (observed): The Prisma Browser deployment docs describe Windows MSI installers for x64 and ARM64 and a macOS installation, with weekly updates.Quote: "Prisma Browser supports two installation variations for macOS environments."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/content/techdocs/en_US/prisma-browser/deployment), read 2026-11-16. - **palo-alto-networks-24** (none-found): None found on 2026-11-16: a published price for Prisma AIRS, AI Access Security or Prisma Browser on their product pages; guessed pricing URLs on paloaltonetworks.com returned 404.Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/sase/ai-access-security), read 2026-11-16. - **palo-alto-networks-26** (vendor): Palo Alto Networks lists a published research report on prompt attacks among the AI Access Security resources.Quote: "Securing GenAI: A Comprehensive Report on Prompt Attacks: Taxonomy, Risks and Solutions"Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/sase/ai-access-security), read 2026-11-16. - **palo-alto-networks-27** (vendor): Palo Alto Networks says Gartner recognized it as a Market Shaper in the Emerging Market Quadrant for AI Application Security in September 2026.The exact quote is on the source page.Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/prisma/prisma-ai-runtime-security), read 2026-11-16. - **palo-alto-networks-28** (vendor): Palo Alto Networks says it is a NASDAQ-listed company with more than 70,000 customers.Quote: "Trusted by 70,000+ customers and powered by Unit 42 threat intelligence"Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/company/press/2026/palo-alto-networks-completes-acquisition-of-portkey-to-secure-ai-agents), read 2026-11-16. - **palo-alto-networks-29** (vendor): Prisma AIRS lists Glean's CISO as a named customer, quoted on the product page.Quote: "What makes Palo Alto Networks AI Runtime stand out are three things."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/prisma/prisma-ai-runtime-security), read 2026-11-16. - **palo-alto-networks-30** (vendor): Prisma AIRS lists Moveworks and TELUS Digital as named customers, quoted on the product page.Quote: "A product like Prisma AIRS is not just good for Moveworks. It's absolutely essential for the industry."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/prisma/prisma-ai-runtime-security), read 2026-11-16. - **palo-alto-networks-31** (vendor): The Prisma Browser page quotes Wipro's CISO as a named customer.Quote: "Prisma Browser delivered true zero trust security to the last mile."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/sase/prisma-browser), read 2026-11-16. - **palo-alto-networks-32** (third-party): On 2026-08-20 SiliconANGLE reported that Zero Networks expanded its Palo Alto Networks integration, with an AI Segmentation link to Prisma AIRS that has no release date.Quote: "The Prisma AIRS integration and Windows traffic redirection has not been given a release date."Source: [siliconangle.com](https://siliconangle.com/2026/08/20/zero-networks-expands-palo-alto-networks-integration-to-ai-agent-control/), read 2026-11-16. - **palo-alto-networks-33** (vendor): Palo Alto Networks documents SaaS Agent Security for agents built on SaaS platforms such as Microsoft Copilot Studio and ServiceNow.Quote: "SaaS Agent Security addresses the unique security problems created by AI agents built on SaaS platforms like Microsoft Copilot Studio and ServiceNow."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-agent-identity/saas-agent-security-overview), read 2026-11-16. - **palo-alto-networks-34** (vendor): The SaaS Agent Security docs describe enforcement workflows that remediate, pause, deactivate or terminate risky agents.Quote: "native, automated enforcement workflows that can remediate, pause, deactivate, or terminate risky agents"Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-agent-identity/saas-agent-security-overview), read 2026-11-16. - **palo-alto-networks-35** (observed): The Prisma AIRS docs describe AI Agent Discovery of agents in an onboarded cloud account, with a console list of enterprise agents.Quote: "AI Agent Discovery allows you to discover agents from an onboarded cloud account and secure them using the AI Runtime API Intercept workflow."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/content/techdocs/en_US/prisma-airs/ai-runtime-security/airs-api/protect-agents-using-api-intercept.html), read 2026-11-16. - **palo-alto-networks-36** (vendor): Palo Alto Networks says SaaS Agent Security shows all agents, their configurations and posture, with auditability and reports.Quote: "SaaS Agent Security provides a comprehensive view of all agents, their configurations, and security postures"Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-agent-identity/saas-agent-security-overview), read 2026-11-16. - **palo-alto-networks-37** (vendor): The Prisma AIRS docs say cloud account onboarding gives continuous discovery of AI and non-AI resources on AWS, Azure and GCP.Quote: "By onboarding your AWS, Azure, or GCP accounts, you enable continuous discovery of both AI and non-AI resources"Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-inventory/cloud-resource-discovery), read 2026-11-16. - **palo-alto-networks-38** (vendor): The Prisma AIRS page lists an October 2026 update that brings unified identity to AI agents acting on behalf of users.Quote: "Agent Identity Security brings unified identity to AI agents acting on behalf of users."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/prisma/prisma-ai-runtime-security), read 2026-11-16. - **palo-alto-networks-39** (vendor): The Prisma AIRS docs say AI Agent Identity secures agents and MCP-based workflows with zero-trust identity governance.Quote: "Secures AI agents and MCP-based workflows with zero-trust identity governance"Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/ai-runtime-security), read 2026-11-16. - **palo-alto-networks-40** (vendor): On 2026-05-29 Palo Alto Networks said agent identity security via Idira authenticates every agentic interaction.Quote: "Authenticating every agentic interaction to prevent unauthorized tool use and lateral movement, helping ensure all agents are treated as privileged users."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/company/press/2026/palo-alto-networks-completes-acquisition-of-portkey-to-secure-ai-agents), read 2026-11-16. - **palo-alto-networks-41** (vendor): The Prisma AIRS docs say AI Gateway sits between users or applications and every LLM provider they call.Quote: "sits between your users and applications and every large language model (LLM) provider they call."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview), read 2026-11-16. - **palo-alto-networks-42** (vendor): The Prisma AIRS docs say AI Gateway guardrails run inline on every request and response, blocking prompt injection and detecting sensitive data.Quote: "Guardrails run inline on every request and response, blocking prompt injection attempts, detecting sensitive data, and enforcing topic restrictions"Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview), read 2026-11-16. - **palo-alto-networks-43** (vendor): The AI Gateway docs say workspaces set LLM access, budgets, rate limits and guardrail policy per team or application.Quote: "logical containers that let you manage LLM access, budgets, rate limits, and guardrail policies per team or application."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview), read 2026-11-16. - **palo-alto-networks-44** (vendor): The AI Gateway docs offer a SaaS model and a Hybrid model where the data plane runs in the customer's own Kubernetes cluster.Quote: "Hybrid , where you host the data plane in your own Kubernetes cluster so that request and response payloads never leave your environment."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview), read 2026-11-16. - **palo-alto-networks-45** (vendor): The Prisma AIRS MCP Server docs say it intercepts tool invocations, scans them and returns a verdict, as a tool the agent is instructed to call.Quote: "The Prisma AIRS MCP Server operates by intercepting tool invocations, performing security analysis, and then returning a verdict on whether a threat was detected."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/content/techdocs/en_US/prisma-airs/ai-agent-identity/understanding-the-prisma-airs-mcp-server.html), read 2026-11-16. - **palo-alto-networks-46** (vendor): The Prisma AIRS MCP Server docs say detailed logs track tool invocations, security verdicts and detected threats.Quote: "You can monitor these interactions through detailed logs that track tool invocations, security verdicts, and any detected threats."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/content/techdocs/en_US/prisma-airs/ai-agent-identity/understanding-the-prisma-airs-mcp-server.html), read 2026-11-16. - **palo-alto-networks-47** (vendor): The AI Security Profile docs say a Block action intercepts and stops the request and records a violation.Quote: "Block: Intercepts and stops the request, records a violation at the configured severity"Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/content/techdocs/en_US/prisma-airs/ai-runtime-security/airs-api/configure-allow-alert-block-actions-ai-security-profiles.html), read 2026-11-16. - **palo-alto-networks-48** (vendor): Palo Alto Networks says Prisma AIRS agent security addresses memory manipulation and tool misuse.Quote: "It provides dedicated AI Agent Security against risks like memory manipulation and tool misuse, ensuring safe interactions with tools and workflows."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/prisma/prisma-ai-runtime-security), read 2026-11-16. - **palo-alto-networks-49** (vendor): Palo Alto Networks lists Secure Agentic Browsing for Prisma Browser, with governance of automated tasks.Quote: "Ensure safe execution and governance of automated tasks."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/sase/prisma-browser), read 2026-11-16. - **palo-alto-networks-50** (observed): A Palo Alto Networks GitHub repo holds community example hooks for coding assistants (Claude Code, Codex, Cursor, Cline, Gemini CLI and others), supported on a best-effort basis.Quote: "supported as best effort by Palo Alto Networks"Source: [github.com](https://github.com/PaloAltoNetworks/prisma-airs-integrations), read 2026-11-16. - **palo-alto-networks-51** (observed): The same GitHub repo says Prisma AIRS scans prompts, responses and tool interactions, and lists gateway and coding assistant examples.Quote: "Prisma AIRS provides inline security for AI applications, scanning prompts, responses, and tool interactions in real-time."Source: [github.com](https://github.com/PaloAltoNetworks/prisma-airs-integrations), read 2026-11-16. - **palo-alto-networks-52** (vendor): Palo Alto Networks says Prisma AIRS offers Network Intercept and API Intercept deployment modes.Quote: "It offers deployment modes such as Network Intercept and API Intercept to secure AI applications, AI models, AI data, and AI agents."Source: [paloaltonetworks.com](https://www.paloaltonetworks.com/prisma/prisma-ai-runtime-security), read 2026-11-16. - **palo-alto-networks-55** (vendor): The AI Runtime Security docs describe an instance deployed in the customer's cloud environment that inspects traffic between AI apps, models and datasets.Quote: "An Prisma AIRS instance is a specific deployment of AI Runtime Security within your cloud environment."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/getting-started/ai-runtime-security-overview), read 2026-11-16. - **palo-alto-networks-66** (vendor): The SaaS Agent Security docs say security teams can open actionable tickets directly from findings.Quote: "enabling security teams to create actionable tickets directly from security findings."Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/prisma-airs/ai-agent-identity/saas-agent-security-overview), read 2026-11-16. - **palo-alto-networks-67** (none-found): None found on 2026-11-16: SIEM or SOAR export, or IdP and EDR integrations, in the Prisma AIRS docs pages read (overview, API Intercept, gateway, SaaS agent security).Source: [docs.paloaltonetworks.com](https://docs.paloaltonetworks.com/ai-runtime-security), read 2026-11-16. - **palo-alto-networks-70** (third-party): On 2025-08-19 The Register reported that Palo Alto Networks owns the Prisma Access Browser (now Prisma Browser) and sells it with its SASE products; this is older than the 90-day window.The exact quote is on the source page.Source: [theregister.com](https://www.theregister.com/security/2025/08/19/browser-wars-are-back-predicts-palo-alto-thanks-to-ai/1228881), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Operant AI profile](/vendors/operant-ai/) - [Pillar Security profile](/vendors/pillar-security/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Pillar Security AI agent security URL: https://aiagentsecuritysolutions.com/vendors/pillar-security/ Updated: 2026-11-16 Pillar Security focuses on coding agents on managed machines. Per its pages, Agentic Endpoint finds coding agents, MCP servers, skills and plugins, scans Claude Code, Cursor, Codex CLI, Gemini CLI and others in depth, and rolls out through MDM scripts on macOS, Linux and Windows. Runtime Guardian sits in the pre-tool hook and decides on every action before it runs, blocking key and credential reads headed outside and persistence git hooks, with managed hooks developers cannot disable. Runtime guardrails work with AI gateways such as LiteLLM and Kong. Browser agents, named SaaS and cloud agent platforms and a price: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Blocks agent actions before they run through pre-tool hooks on managed machines, with a gateway integration list (pillar-security-09, pillar-security-27). **Weakness:** None found on 2026-11-16 for browser agents or named SaaS and cloud agent platforms (pillar-security-45, pillar-security-46). ## Scores and evidence Pillar Security is 14th in the overall ranking at 53.1 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Pillar Security: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 7 of 10 | 6 | The vendor names the coding agents it scans and sits in their pre-tool hooks with MDM-managed hooks; vendor claims only. | [pillar-security-04](#fact-pillar-security-04), [pillar-security-05](#fact-pillar-security-05), [pillar-security-09](#fact-pillar-security-09), [pillar-security-11](#fact-pillar-security-11) | | Coverage: browser layer | 0 of 10 | 6 | None found: browser agents, AI browsers or extensions are not covered. | [pillar-security-45](#fact-pillar-security-45), [pillar-security-49](#fact-pillar-security-49) | | Coverage: cloud workspace layer | 2 of 10 | 6 | The vendor identifies external AI services including SaaS; named Copilot Studio, Agentforce, ServiceNow or Bedrock coverage none found. | [pillar-security-22](#fact-pillar-security-22), [pillar-security-46](#fact-pillar-security-46) | | Coverage: identity layer | 2 of 10 | 6 | Posture issues include identity and RBAC for AI assets; agent credential inventory, IdP integration and revocation none found. | [pillar-security-16](#fact-pillar-security-16), [pillar-security-23](#fact-pillar-security-23), [pillar-security-46](#fact-pillar-security-46), [pillar-security-47](#fact-pillar-security-47) | | Coverage: gateway layer | 5 of 10 | 6 | Runtime guardrails work with named AI gateways via API; the vendor does not document its own gateway or MCP proxy. | [pillar-security-25](#fact-pillar-security-25), [pillar-security-27](#fact-pillar-security-27) | | Stops the action | 7 of 10 | 20 | The vendor blocks credential reads, exfiltration, persistence git hooks and config rewrites before actions run; vendor claims only. | [pillar-security-09](#fact-pillar-security-09), [pillar-security-10](#fact-pillar-security-10), [pillar-security-11](#fact-pillar-security-11) | | Visibility | 7 of 10 | 15 | The vendor catalogs agents, tools and MCP servers, audits permissions and logs prompts and tool calls; vendor claims only. | [pillar-security-07](#fact-pillar-security-07), [pillar-security-08](#fact-pillar-security-08), [pillar-security-17](#fact-pillar-security-17), [pillar-security-28](#fact-pillar-security-28) | | Deployment | 6 of 10 | 10 | MDM scripts on three OSes and cloud, self-hosted or hybrid options are documented; a time to deploy none found. | [pillar-security-06](#fact-pillar-security-06), [pillar-security-29](#fact-pillar-security-29), [pillar-security-30](#fact-pillar-security-30), [pillar-security-33](#fact-pillar-security-33) | | Policy and governance | 7 of 10 | 10 | Off, Monitor or Block controls, audit trail, SIEM streaming and ticketing integration are described; IdP integration none found. | [pillar-security-12](#fact-pillar-security-12), [pillar-security-19](#fact-pillar-security-19), [pillar-security-21](#fact-pillar-security-21), [pillar-security-24](#fact-pillar-security-24), [pillar-security-28](#fact-pillar-security-28) | | Transparency | 3 of 10 | 10 | Research and a framework are published and covered by a third party; public documentation and a price none found. | [pillar-security-34](#fact-pillar-security-34), [pillar-security-35](#fact-pillar-security-35), [pillar-security-36](#fact-pillar-security-36), [pillar-security-43](#fact-pillar-security-43) | | Proof | 6 of 10 | 5 | A third party reports a $9 million seed and a Fortune 300 proof of concept, with named testimonials; independent analyst reports are vendor-cited only. | [pillar-security-37](#fact-pillar-security-37), [pillar-security-39](#fact-pillar-security-39), [pillar-security-40](#fact-pillar-security-40), [pillar-security-41](#fact-pillar-security-41) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 14th | 53.1 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | joint 5th | 64.5 | | [Ranked for the browser layer](/best-for/layer-browser/) | joint 14th | 43.5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 13th | 49.5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 12th | 49.5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 12th | 58.5 | ## Product, deployment, price and owner - Product assessed: Agentic Endpoint, Runtime Guardrails and AI Discovery & Posture - Deployment: Managed cloud, self-hosted or hybrid; endpoint product by MDM script, with a per-user service and agent hooks - Time to deploy: none found on 2026-11-16 - Price: none found on 2026-11-16 - Owner: Pillar Security. Vanta announced a definitive agreement to acquire it on 9 Nov 2026; the deal was not stated as closed on 16 Nov 2026 - Website: [pillar.security](https://www.pillar.security) ## Facts and sources - **pillar-security-01** (vendor): On 2026-11-09 Pillar's founders announced that Vanta has signed a definitive agreement to acquire Pillar Security, so the deal is announced and not described as closed.Quote: "Today, Vanta has signed a definitive agreement to acquire Pillar Security."Source: [pillar.security](https://www.pillar.security/blog/vanta-to-acquire-pillar-security-building-trust-in-ai-together), read 2026-11-16. - **pillar-security-02** (vendor): Pillar says it was founded in 2023 and describes its product as discovering, testing and securing AI applications and agents and blocking unauthorized actions at runtime.Quote: "When we founded Pillar in 2023, most of the conversation around AI security was about chatbots and prompts."Source: [pillar.security](https://www.pillar.security/blog/vanta-to-acquire-pillar-security-building-trust-in-ai-together), read 2026-11-16. - **pillar-security-03** (vendor): Pillar says its platform discovers, tests, secures and governs AI agents, across five modules: AI Discovery & Posture, Red Teaming & Attack Surface Exposure, Runtime Guardrails, Agentic Endpoint, and Governance & Compliance.Quote: "One platform to discover, test, secure and govern every AI agent across your organization"Source: [pillar.security](https://www.pillar.security), read 2026-11-16. - **pillar-security-04** (vendor): Pillar says its Agentic Endpoint product finds coding agents, MCP servers, skills and plugins on every managed machine.Quote: "Find coding agents, MCP servers, skills and plugins on every machine."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-05** (vendor): Pillar names the agents it scans in depth (Claude Code, Cursor, Codex CLI, Gemini CLI, OpenCode, Pi) and others it detects by fingerprint (Amazon Q, Antigravity, OpenClaw).Quote: "Deep scans for Claude Code, Cursor, Codex CLI, Gemini CLI, OpenCode and Pi; fingerprinting for Amazon Q, Antigravity, OpenClaw and more."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-06** (vendor): Pillar says the endpoint product is deployed through MDM scripts (Jamf Pro, Kandji, JumpCloud or any MDM that runs scripts) on macOS, Linux and Windows.Quote: "Roll out with Jamf Pro, Kandji, JumpCloud or any MDM that runs scripts, on macOS, Linux and Windows."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-07** (vendor): Pillar says its endpoint product flags third-party agents the organization has not approved.Quote: "Flag third-party agents your organization has not approved."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-08** (vendor): Pillar says it audits agent configuration such as permission rules, approval and sandbox modes, MCP servers, hooks and plugins, and maps issues to its SAIL framework.Quote: "Pillar parses each agent's permission rules, approval and sandbox modes, MCP servers, hooks and plugins"Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-09** (vendor): Pillar says Runtime Guardian sits in the pre-tool hook of Claude Code, Codex CLI and Cursor and decides on every action before it runs.Quote: "Runtime Guardian sits in the agent's own pre-tool hook for Claude Code, Codex CLI and Cursor, and decides on every action before it runs."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-10** (vendor): Pillar says it blocks reads of keys, .env, SSH and cloud credentials from reaching external destinations, and blocks persistence git hooks and agents rewriting their own configuration.Quote: "Stop reads of keys, .env, SSH and cloud credentials from reaching external destinations."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-11** (vendor): Pillar says endpoint decisions run locally and keep working offline, and that on macOS and Linux MDM installs managed hooks for Codex and Cursor that developers cannot disable.Quote: "Decisions run locally and keep working offline; on macOS and Linux, MDM installs managed hooks for Codex and Cursor that developers cannot disable."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-12** (vendor): Pillar says each control can be set to Off, Monitor or Block, and every blocked or asked decision appears in an Activity view with the full session timeline.Quote: "Every decision, blocked or asked, appears in Activity with the full session timeline."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-14** (vendor): Pillar says each MCP risk score covers reputation, supply chain, prompt handling, data access and actions, and per-server allow-lists restrict which tools agents may call, in Monitor or Block mode.Quote: "Restrict which tools each MCP server may invoke, in Monitor or Block mode."Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-16** (vendor): Pillar says its agent posture issues are grouped across permissions, code execution and identity, with severity the customer controls.Quote: "turns risky settings into prioritized issues across permissions, code execution and identity"Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-17** (vendor): Pillar says its discovery module catalogs agents, models, prompts, tools, MCP servers and coding agents through agentless integrations with source code repos, data platforms and endpoints.Quote: "Pillar catalogs agents, models, prompts, tools, MCP servers, and coding agents through agentless integrations with source code repos, data platforms, and endpoints."Source: [pillar.security](https://www.pillar.security/platform/ai-discovery-posture), read 2026-11-16. - **pillar-security-18** (vendor): Pillar says it detects shadow AI by identifying unauthorized models, coding agents and MCP servers.Quote: "Identify unauthorized models, coding agents, and MCP servers."Source: [pillar.security](https://www.pillar.security/platform/ai-discovery-posture), read 2026-11-16. - **pillar-security-19** (vendor): Pillar says policy rules can block and track unapproved models and stop them from deploying, with real-time alerts and automated remediation workflows.Quote: "Block and track unapproved models and stop them from deploying."Source: [pillar.security](https://www.pillar.security/platform/ai-discovery-posture), read 2026-11-16. - **pillar-security-20** (vendor): Pillar says it generates audit reports mapped to GDPR, SOC 2, the EU AI Act and ISO 42001.Quote: "Generate audit reports for GDPR, SOC 2, EU AI Act, and ISO 42001."Source: [pillar.security](https://www.pillar.security/platform/ai-discovery-posture), read 2026-11-16. - **pillar-security-21** (vendor): Pillar says posture tracking integrates with ticketing and operational systems to flag critical issues.Quote: "Integrates with ticketing and operational systems to flag critical issues."Source: [pillar.security](https://www.pillar.security/platform/ai-discovery-posture), read 2026-11-16. - **pillar-security-22** (vendor): Pillar says its governance module identifies external AI services, APIs and integrations across the organization, including SaaS platforms.Quote: "Identify external AI services, APIs, and integrations across your organization."Source: [pillar.security](https://www.pillar.security/platform/governance-compliance), read 2026-11-16. - **pillar-security-23** (vendor): Pillar says governance includes RBAC for AI assets controlling who can deploy models, access sensitive data and modify policies.Quote: "Control who can deploy models, access sensitive data, and modify policies."Source: [pillar.security](https://www.pillar.security/platform/governance-compliance), read 2026-11-16. - **pillar-security-24** (vendor): Pillar says its audit trail logs every model deployment, policy decision and compliance check with metadata.Quote: "Log every model deployment, policy decision, and compliance check with complete metadata."Source: [pillar.security](https://www.pillar.security/platform/governance-compliance), read 2026-11-16. - **pillar-security-25** (vendor): Pillar says its runtime guardrails detect and block prompt injection, jailbreaking and tool manipulation attempts.Quote: "Detect and block prompt injection, jailbreaking, and tool manipulation attempts."Source: [pillar.security](https://www.pillar.security/platform/runtime-guardrails), read 2026-11-16. - **pillar-security-27** (vendor): Pillar lists AI gateways it works with (LiteLLM, Kong, TrueFoundry, Agent Router, Open WebUI) and says any gateway works via API.Quote: "+ any gateway via API"Source: [pillar.security](https://www.pillar.security/platform/runtime-guardrails), read 2026-11-16. - **pillar-security-28** (vendor): Pillar says it logs prompts, responses, tool calls and results with timestamps, and streams telemetry to SIEM platforms.Quote: "Stream telemetry to security platforms for centralized monitoring."Source: [pillar.security](https://www.pillar.security/platform/runtime-guardrails), read 2026-11-16. - **pillar-security-29** (vendor): Pillar says it runs as managed cloud, self-hosted in the customer's cloud account or data center, or hybrid.Quote: "Deploy Pillar where your data lives: our managed cloud, your own cloud account or data center, or a mix of both."Source: [pillar.security](https://www.pillar.security/platform/deployment), read 2026-11-16. - **pillar-security-30** (vendor): Pillar's go-live steps are scope, install, connect (AI gateway, code repositories, device management) and go live in monitor mode before blocking.Quote: "Start in monitor mode, then switch to blocking."Source: [pillar.security](https://www.pillar.security/platform/deployment), read 2026-11-16. - **pillar-security-33** (none-found): None found on 2026-11-16: a published time to deploy or time to first inventory, searched on the deployment, agentic endpoint and discovery pages.Source: [pillar.security](https://www.pillar.security/platform/deployment), read 2026-11-16. - **pillar-security-34** (none-found): None found on 2026-11-16: a published price; the pricing URL returned 404 and the site offers only a demo request.Source: [pillar.security](https://www.pillar.security/pricing), read 2026-11-16. - **pillar-security-35** (none-found): None found on 2026-11-16: public product documentation; docs.pillar.security loads a sign-in page marked noindex.Source: [docs.pillar.security](https://docs.pillar.security), read 2026-11-16. - **pillar-security-36** (vendor): Pillar publishes the SAIL framework and a research blog and says SAIL has passed 50,000 downloads and maps 91 AI risks across seven lifecycle phases.Quote: "SAIL has passed 50,000 downloads, and maps 91 AI risks across seven lifecycle phases."Source: [pillar.security](https://www.pillar.security/blog/vanta-to-acquire-pillar-security-building-trust-in-ai-together), read 2026-11-16. - **pillar-security-37** (vendor): Pillar's site shows testimonials from named people at Eleos, another company and Similarweb.The exact quote is on the source page.Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. - **pillar-security-38** (vendor): Pillar says a Fortune 300 manufacturer runs it fully self-hosted.Quote: "A Fortune 300 manufacturer runs Pillar fully self-hosted."Source: [pillar.security](https://www.pillar.security/platform/deployment), read 2026-11-16. - **pillar-security-39** (third-party): CRN reported that Pillar announced a $9 million seed round led by Shield Capital in April 2025, and listed it among its 10 hottest AI security startups of 2026.Quote: "Pillar announced its $9 million seed funding round led by Shield Capital in April 2025."Source: [crn.com](https://www.crn.com/news/security/2026/the-10-hottest-ai-security-startups-of-2026-so-far?page=6), read 2026-11-16. - **pillar-security-40** (vendor): The Pillar newsroom lists a Gartner Cool Vendor in AI Software Security recognition dated Jul 1, 2026 and a Frost & Sullivan award dated Sep 17, 2025.Quote: "Pillar Security Named as a 2026 Gartner® Cool Vendor in AI Software Security"Source: [pillar.security](https://www.pillar.security/newsroom), read 2026-11-16. - **pillar-security-41** (third-party): A Refresh Miami profile dated 2026-01-26 says Pillar scanned more than 30,000 repositories in a Fortune 300 proof of concept and found 8,000 AI models and 400 AI systems unknown to the security team.Quote: "Pillar scanned more than 30,000 code repositories. The result: 8,000 AI models and 400 AI systems"Source: [refreshmiami.com](https://refreshmiami.com/news/ai-broke-the-security-model-pillar-wants-to-fix-it/), read 2026-11-16. - **pillar-security-43** (third-party): SecurityWeek reported on 2026-05-07 that Pillar disclosed a critical Gemini CLI vulnerability that could have led to supply chain compromise via a GitHub issue.Quote: "A critical vulnerability in Gemini CLI could have allowed attackers to mount a supply chain attack via indirect prompts injected into a GitHub issue, Pillar Security warns."Source: [securityweek.com](https://www.securityweek.com/gemini-cli-vulnerability-could-have-led-to-code-execution-supply-chain-attack/), read 2026-11-16. - **pillar-security-44** (vendor): The Pillar newsroom lists a press release dated Oct 13, 2026 announcing BadEngram research on supply-chain risk in open-weight AI models.Quote: "Pillar Security Discovers ‘BadEngram,’ Exposing Hidden Supply-Chain Risk in Open-Weight AI Models"Source: [pillar.security](https://www.pillar.security/newsroom), read 2026-11-16. - **pillar-security-45** (none-found): None found on 2026-11-16: browser agent, AI browser or browser extension discovery or control, searched on the platform, discovery and agentic endpoint pages.Source: [pillar.security](https://www.pillar.security/platform/ai-discovery-posture), read 2026-11-16. - **pillar-security-46** (none-found): None found on 2026-11-16: named coverage of Microsoft Copilot Studio, Salesforce Agentforce, ServiceNow or Bedrock agents, or agent token and OAuth grant revocation, on the platform pages.Source: [pillar.security](https://www.pillar.security/platform/governance-compliance), read 2026-11-16. - **pillar-security-47** (none-found): None found on 2026-11-16: named identity provider (Okta, Entra ID), OAuth-grant, proxy or network-traffic integrations for AI app discovery, on the discovery and governance pages.Source: [pillar.security](https://www.pillar.security/platform/ai-discovery-posture), read 2026-11-16. - **pillar-security-49** (none-found): None found on 2026-11-16: browser agents, AI browsers or AI extensions in the list of supported agents, which names only Claude Code, Cursor, Codex CLI, Gemini CLI, OpenCode and Pi.Source: [pillar.security](https://www.pillar.security/platform/agentic-endpoint), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Palo Alto Networks profile](/vendors/palo-alto-networks/) - [SentinelOne (Prompt Security) profile](/vendors/sentinelone/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # SentinelOne AI agent security (Prompt Security) URL: https://aiagentsecuritysolutions.com/vendors/sentinelone/ Updated: 2026-11-16 SentinelOne sells agent security through Prompt Security. Its pages describe local sensors, a browser extension and a dedicated AI endpoint agent, plus an MCP Gateway and API integrations, with an inventory of agents, MCP servers and connectors and a session and tool log. A press release describes an integration with Amazon Bedrock AgentCore where AgentCore acts on Prompt Security signals at the gateway, outside the agent's reasoning loop, and parts are marked upcoming. Research and open-source code are public. Copilot Studio, Agentforce and ServiceNow coverage, agent credential and token controls, a product price and public documentation: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers endpoint, browser, gateway and Bedrock AgentCore layers with an agent and MCP inventory and audit log (sentinelone-13, sentinelone-14, sentinelone-35, sentinelone-44). **Weakness:** None found on 2026-11-16 for Copilot Studio, Agentforce or ServiceNow coverage, or agent credential and token controls (sentinelone-37, sentinelone-39). ## Scores and evidence SentinelOne (Prompt Security) is 16th in the overall ranking at 46.8 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **SentinelOne (Prompt Security): score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 5 of 10 | 6 | The vendor tracks named code assistants with an AI endpoint agent and publishes an agent security skill suite; endpoint enforcement detail is mostly stated as goals. | [sentinelone-09](#fact-sentinelone-09), [sentinelone-13](#fact-sentinelone-13), [sentinelone-32](#fact-sentinelone-32), [sentinelone-28](#fact-sentinelone-28) | | Coverage: browser layer | 4 of 10 | 6 | A browser extension governs AI use across browsers; browser agent specifics none found. | [sentinelone-11](#fact-sentinelone-11), [sentinelone-13](#fact-sentinelone-13) | | Coverage: cloud workspace layer | 3 of 10 | 6 | Only Bedrock AgentCore is named and parts are upcoming; other SaaS agent platforms none found. | [sentinelone-34](#fact-sentinelone-34), [sentinelone-35](#fact-sentinelone-35), [sentinelone-36](#fact-sentinelone-36), [sentinelone-37](#fact-sentinelone-37) | | Coverage: identity layer | 2 of 10 | 6 | Least-privilege scope is claimed; credential inventory and revocation none found. | [sentinelone-38](#fact-sentinelone-38), [sentinelone-39](#fact-sentinelone-39) | | Coverage: gateway layer | 4 of 10 | 6 | An MCP gateway is named and a third party mentions it, with few specifics; the AgentCore gateway enforcement is an upcoming integration. | [sentinelone-14](#fact-sentinelone-14), [sentinelone-34](#fact-sentinelone-34), [sentinelone-35](#fact-sentinelone-35), [sentinelone-40](#fact-sentinelone-40) | | Stops the action | 5 of 10 | 20 | Prompt blocking and allow-or-deny at the AgentCore gateway are described; named agent action types are few. | [sentinelone-35](#fact-sentinelone-35), [sentinelone-41](#fact-sentinelone-41), [sentinelone-42](#fact-sentinelone-42), [sentinelone-43](#fact-sentinelone-43) | | Visibility | 7 of 10 | 15 | Inventory of agents, MCP servers and connectors plus a session and tool log is described; vendor claims only. | [sentinelone-07](#fact-sentinelone-07), [sentinelone-10](#fact-sentinelone-10), [sentinelone-44](#fact-sentinelone-44), [sentinelone-45](#fact-sentinelone-45) | | Deployment | 5 of 10 | 10 | Extension, endpoint agent and on-prem options deploying in minutes are claimed; OS lists and requirements none found. | [sentinelone-17](#fact-sentinelone-17), [sentinelone-46](#fact-sentinelone-46), [sentinelone-55](#fact-sentinelone-55), [sentinelone-18](#fact-sentinelone-18) | | Policy and governance | 4 of 10 | 10 | Audit trails and risk scoring exist; named SIEM, EDR, IdP or ticketing integrations none found. | [sentinelone-22](#fact-sentinelone-22), [sentinelone-36](#fact-sentinelone-36), [sentinelone-45](#fact-sentinelone-45), [sentinelone-48](#fact-sentinelone-48) | | Transparency | 3 of 10 | 10 | Research and open-source code are public; a product price and public documentation none found. | [sentinelone-25](#fact-sentinelone-25), [sentinelone-26](#fact-sentinelone-26), [sentinelone-27](#fact-sentinelone-27), [sentinelone-28](#fact-sentinelone-28) | | Proof | 7 of 10 | 5 | Third parties report the funding and acquisition, with named CISO quotes; analyst coverage is vendor-cited. | [sentinelone-03](#fact-sentinelone-03), [sentinelone-04](#fact-sentinelone-04), [sentinelone-05](#fact-sentinelone-05), [sentinelone-29](#fact-sentinelone-29) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 16th | 46.8 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 16th | 51.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 11th | 48.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 15th | 45.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 16th | 42.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 15th | 48.0 | ## Product, deployment, price and owner - Product assessed: Prompt Security Agentic AI Security, with the MCP Gateway and Amazon Bedrock AgentCore integration (SentinelOne) - Deployment: Singularity Platform console; browser extension, dedicated AI endpoint agent, MCP Gateway and API integrations; self-hosted Prompt Security On-Premises; AgentCore gateway integration - Time to deploy: SentinelOne says it deploys in minutes; no measured figure or OS list found on 2026-11-16 - Price: none found on 2026-11-16 for Prompt Security; the pricing page lists Singularity tiers and says Enterprise is Contact Sales for Pricing - Owner: SentinelOne, Inc. (NYSE: S), Mountain View, California - Website: sentinelone.com ## Facts and sources - **sentinelone-01** (vendor): SentinelOne says Prompt Security is built into its Singularity Platform and managed from the same console as its endpoint, cloud, identity and data security.Quote: "Prompt Security is built into the SentinelOne Singularity Platform."Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-02** (vendor): On 2025-08-05 SentinelOne announced a definitive agreement to acquire Prompt Security, so the AI security line sits inside SentinelOne (NYSE: S).Quote: "signed a definitive agreement to acquire Prompt Security"Source: [sentinelone.com](https://www.sentinelone.com/press/sentinelone-to-acquire-prompt-security-to-advance-genai-security/), read 2026-11-16. - **sentinelone-03** (third-party): CRN reports that Prompt Security was founded in 2023 and raised an $18 million Series A in November 2024 before SentinelOne agreed to acquire it.Quote: "Founded in 2023, Prompt Security offers protection against AI-related threats"Source: [crn.com](https://www.crn.com/news/security/2025/sentinelone-to-acquire-ai-protection-startup-prompt-security/), read 2026-11-16. - **sentinelone-04** (third-party): CRN reports the Series A was led by Jump Capital with Hetz Ventures, Ridge Ventures, Okta and F5 taking part.Quote: "In November 2024, Prompt Security announced an $18 million Series A funding round led by Jump Capital"Source: [crn.com](https://www.crn.com/news/security/2025/sentinelone-to-acquire-ai-protection-startup-prompt-security/), read 2026-11-16. - **sentinelone-05** (third-party): Techzine reports SentinelOne acquired Israeli company Prompt Security to prepare its Singularity platform for secure AI use in organizations.Quote: "SentinelOne has announced the acquisition of Israeli company Prompt Security."Source: [techzine.eu](https://www.techzine.eu/news/security/133547/sentinelone-strengthens-ai-protection-with-prompt-security-deal/), read 2026-11-16. - **sentinelone-06** (third-party): Wikipedia lists SentinelOne as a public company headquartered in Mountain View, California, founded in 2013, with Tomer Weingarten as chairman and CEO.Quote: "Headquarters Mountain View, California , U.S."Source: [en.wikipedia.org](https://en.wikipedia.org/wiki/SentinelOne), read 2026-11-16. - **sentinelone-07** (vendor): SentinelOne says Prompt Security inventories every AI tool and code assistant in use, including unsanctioned shadow AI.Quote: "Inventory every AI tool and code assistant in use, including unsanctioned shadow AI"Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-09** (vendor): SentinelOne says Prompt Security tracks AI code assistants in the environment and names GitHub Copilot, Cursor and Claude Code.Quote: "Track every AI code assistant in the environment, including GitHub Copilot, Cursor, and Claude Code"Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-ai-usage-control/), read 2026-11-16. - **sentinelone-10** (vendor): SentinelOne says Prompt Security discovers every AI agent and MCP server in the environment, including the connectors, tools, skills and plugins an agent touches.Quote: "See every connector, tool, skill, and plugin an agent touches, and who's running it"Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-11** (vendor): SentinelOne says AI Usage Control discovers and governs AI use across employees, browsers and embedded copilots, unsanctioned tools included.Quote: "Discover and govern AI usage across every employee, browser, and embedded copilot, unsanctioned tools included."Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-ai-usage-control/), read 2026-11-16. - **sentinelone-13** (vendor): Prompt Security says AI governance needs local sensors, namely a browser extension and a dedicated AI endpoint agent, because network controls alone fall short.Quote: "you need local, purpose-built sensors: a browser extension and a dedicated AI endpoint agent"Source: [prompt.security](https://prompt.security/blog/the-key-layer-in-ai-security-browser-and-endpoint-sensors), read 2026-11-16. - **sentinelone-14** (vendor): SentinelOne says an MCP Gateway, browser extension, security plugins and API integrations work together so no agent surface goes undetected.Quote: "MCP Gateway, browser extension, security plugins, and API integrations work together so no agent surface goes undetected"Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-17** (vendor): SentinelOne says Prompt Security deploys in minutes with no complex integrations or infrastructure changes.Quote: "Prompt Security deploys in minutes. No complex integrations or infrastructure changes are required."Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-18** (none-found): None found on 2026-11-16: looked on the Prompt Security product pages for supported operating systems, browsers, a measured time to first AI inventory and a published deployment requirements list; only "minutes" was found.Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-22** (vendor): SentinelOne says AI Usage Control keeps a full audit trail of every prompt, redaction and policy action.Quote: "Review every prompt, redaction, and policy action in one place, with a full audit trail across every AI interaction"Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-ai-usage-control/), read 2026-11-16. - **sentinelone-25** (vendor): SentinelOne's public pricing page lists tiers but says the Enterprise tier is "Contact Sales for Pricing" and does not list Prompt Security.Quote: "Contact Sales for Pricing"Source: [sentinelone.com](https://www.sentinelone.com/pricing/), read 2026-11-16. - **sentinelone-26** (none-found): None found on 2026-11-16: looked on the pricing page and Prompt Security pages for a published price of Prompt Security or AI Usage Control and for public technical documentation readable without sign-in; found marketing pages and a demo form only.Source: [sentinelone.com](https://www.sentinelone.com/pricing/), read 2026-11-16. - **sentinelone-27** (vendor): On 2026-08-20 Prompt Security published research on three agent cases (AGENTS.md goal hijack, marketplace skills dependency hijack, instruction file integrity) and says approval prompts should show the real destination and payload.Quote: "approval prompts that show the real destination and payload instead of a friendly summary"Source: [prompt.security](https://prompt.security/blog/when-ai-agents-become-the-supply-chain-hidden-control-planes-in-agentic-systems), read 2026-11-16. - **sentinelone-28** (observed): Prompt Security publishes ClawSec on GitHub under AGPL-3.0, a security skill suite for OpenClaw, Hermes, PicoClaw and NanoClaw agents with 1.1k stars.Quote: "ClawSec source code is licensed under GNU AGPL-3.0-or-later"Source: [github.com](https://github.com/prompt-security/clawsec), read 2026-11-16. - **sentinelone-29** (vendor): SentinelOne's page quotes Margarita Rivera, Global CISO of Carnival Corporation, on Prompt Security giving a first layer of visibility and protection.Quote: "Prompt Security has been incredible for our organization."Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-ai-usage-control/), read 2026-11-16. - **sentinelone-30** (vendor): SentinelOne's page quotes Richard Moore, CISO at 10x Banking, on Prompt Security.Quote: "Prompt Security has allowed us to be enablers for the business."Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-ai-usage-control/), read 2026-11-16. - **sentinelone-31** (vendor): SentinelOne's Prompt Security page links a "2026 Latio AI Security Market Report" among its resources (analyst-style coverage cited by the vendor; contents not read).Quote: "2026 Latio AI Security Market Report"Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-32** (vendor): Prompt Security says a dedicated AI endpoint agent should detect new agents, inspect prompt construction, enforce runtime policies and prevent sensitive data exposure before execution.Quote: "Prevent sensitive data exposure before execution"Source: [prompt.security](https://prompt.security/blog/the-key-layer-in-ai-security-browser-and-endpoint-sensors), read 2026-11-16. - **sentinelone-33** (vendor): Prompt Security says monitoring and controlling local MCP communications requires endpoint-level visibility and enforcement.Quote: "Monitoring and controlling MCP communications requires endpoint-level visibility and enforcement."Source: [prompt.security](https://prompt.security/blog/the-key-layer-in-ai-security-browser-and-endpoint-sensors), read 2026-11-16. - **sentinelone-34** (vendor): On 2026-06-17 SentinelOne announced an upcoming Prompt Security integration with Amazon Bedrock AgentCore supplying signals for prompt injection, PII exposure, tool-use validation, LLM response monitoring and data leakage prevention.Quote: "prompt injection detection, PII exposure, tool-use validation, LLM response monitoring, and data leakage prevention"Source: [sentinelone.com](https://www.sentinelone.com/press/sentinelone-announces-ai-security-integration-with-amazon-bedrock-agentcore-to-deliver-runtime-guardrails-for-ai-agents/), read 2026-11-16. - **sentinelone-35** (vendor): SentinelOne says AgentCore acts on Prompt Security signals at the gateway, outside the agent's reasoning loop, with deterministic allow-or-deny enforcement.Quote: "Because detection can be probabilistic, but enforcement stays deterministic"Source: [sentinelone.com](https://www.sentinelone.com/press/sentinelone-announces-ai-security-integration-with-amazon-bedrock-agentcore-to-deliver-runtime-guardrails-for-ai-agents/), read 2026-11-16. - **sentinelone-36** (vendor): On 2026-08-04 SentinelOne said it would connect Prompt Security, Singularity Cloud Security and Singularity AI SIEM in a single view across Amazon Bedrock and AgentCore, with general availability targeted for AWS re:Invent 2026.Quote: "Prompt Security, Singularity™ Cloud Security, and Singularity™ AI SIEM into a single view across Amazon Bedrock"Source: [sentinelone.com](https://www.sentinelone.com/press/sentinelone-expands-collaboration-with-aws-to-deliver-unified-ai-governance/), read 2026-11-16. - **sentinelone-37** (none-found): None found on 2026-11-16: looked on the Agentic AI Security page and SentinelOne press releases for coverage of Microsoft Copilot Studio, Salesforce Agentforce or ServiceNow agents; only Amazon Bedrock AgentCore was named.Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-38** (vendor): SentinelOne says Prompt Security enforces least-privilege access so agents operate only within their defined scope.Quote: "Enforce least-privilege access so agents operate only within their defined scope"Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-39** (none-found): None found on 2026-11-16: looked on the Agentic AI Security page for inventory or revocation of agent credentials and tokens and for named identity provider, EDR or ticketing integrations; none named (a Singularity Identity product is listed separately on the platform page).Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-40** (third-party): Techzine reports Prompt Security works standalone of the AI model and via the Model Context Protocol through MCP gateway security.Quote: "also works via the now established Model Context Protocol via MCP gateway security"Source: [techzine.eu](https://www.techzine.eu/news/security/133547/sentinelone-strengthens-ai-protection-with-prompt-security-deal/), read 2026-11-16. - **sentinelone-41** (vendor): SentinelOne says Prompt Security blocks adversarial prompts and scrubs sensitive outputs with a real-time AI firewall for AI apps.Quote: "Block adversarial prompts and scrub sensitive outputs with a real-time AI firewall"Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-42** (vendor): SentinelOne says Prompt Security enforces policy at the point of interaction, not after the fact.Quote: "Policies enforced at the point of interaction, not after the fact."Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-43** (vendor): SentinelOne says Prompt Security stops agent policy violations before they propagate across connected systems.Quote: "Stop violations before they propagate across connected systems"Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-44** (vendor): SentinelOne says Prompt Security produces a searchable log of every session, action and tool an agent touches.Quote: "Produce a searchable log of every session, action, and tool an agent touches."Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-45** (vendor): SentinelOne says Prompt Security scores every agent and MCP server for risk against the customer's policy.Quote: "Score every agent and MCP server for risk against your policy"Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-46** (vendor): On 2026-03-23 SentinelOne said Prompt Security On-Premises is self-hosted, works in disconnected environments, and blocks prompt injections and data leaks as a firewall for internal and external AI tools.Quote: "Prompt Security blocks threats like prompt injections and data leaks"Source: [sentinelone.com](https://www.sentinelone.com/press/sentinelone-brings-ai-security-to-on-premise-regulated-sovereign-self-hosted-and-airgapped-environments/), read 2026-11-16. - **sentinelone-47** (vendor): SentinelOne says customers with existing Prompt Security licenses can use them with AgentCore through a Bring Your Own License path.Quote: "Customers with existing Prompt Security licenses will apply them through a Bring Your Own License path."Source: [sentinelone.com](https://www.sentinelone.com/press/sentinelone-announces-ai-security-integration-with-amazon-bedrock-agentcore-to-deliver-runtime-guardrails-for-ai-agents/), read 2026-11-16. - **sentinelone-48** (none-found): None found on 2026-11-16: looked on the Prompt Security pages and press releases for a named SIEM export (beyond Singularity AI SIEM), SOAR, EDR, identity provider or ticketing integration for Prompt Security events; none named.Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. - **sentinelone-54** (vendor): SentinelOne says Prompt Security enforces policy at the same speed agents act, but gives no millisecond figure.Quote: "Enforce policy at the same speed agents act"Source: [sentinelone.com](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/), read 2026-11-16. - **sentinelone-55** (vendor): SentinelOne says setup is to connect your environment and integrate with existing infrastructure in minutes, with protection starting immediately.Quote: "Integrate with your existing infrastructure in minutes. No disruption. Protection starts immediately."Source: [sentinelone.com](https://www.sentinelone.com/platform/securing-ai-prompt/), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Pillar Security profile](/vendors/pillar-security/) - [Straiker profile](/vendors/straiker/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Straiker AI agent security URL: https://aiagentsecuritysolutions.com/vendors/straiker/ Updated: 2026-11-16 Straiker documents coverage from coding agents and agentic browsers to SaaS copilots and cloud agent platforms, and its pages name Microsoft Copilot Studio, Salesforce Agentforce, Bedrock AgentCore and Foundry. Browser guardrails govern forms, uploads, downloads, clipboard and screenshots with step approvals, and a kill switch cuts tool access, suspends a session or takes a whole agent offline. It maps OAuth scopes to flag over-permissioned agents and can revoke access. A marketplace listing confirms AgentCore integration. Its own gateway is not documented, only integrations with AI gateways. Public documentation, a price and named SIEM, EDR or ticketing integrations: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers all five layers in documentation, from coding agents and browsers to SaaS copilots, cloud agent platforms and gateways, with a kill switch that stops a whole agent (straiker-06, straiker-12, straiker-24). **Weakness:** No public documentation (docs are behind an access code), no published price, and no named SIEM, EDR, identity or ticketing integrations: none found on 2026-11-16 (straiker-32, straiker-33). ## Scores and evidence Straiker is 2nd in the overall ranking at 63.7 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Straiker: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 7 of 10 | 6 | The vendor names coding agents covered by a thin client and blocks exfiltration, code execution and destructive actions; vendor claims only. | [straiker-06](#fact-straiker-06), [straiker-07](#fact-straiker-07), [straiker-08](#fact-straiker-08) | | Coverage: browser layer | 7 of 10 | 6 | Agentic-browser guardrails govern forms, uploads, downloads, clipboard and screenshots with step approvals; vendor claims only. | [straiker-09](#fact-straiker-09), [straiker-10](#fact-straiker-10), [straiker-11](#fact-straiker-11) | | Coverage: cloud workspace layer | 8 of 10 | 6 | The vendor names Copilot Studio, Agentforce, Bedrock AgentCore and Foundry, and the observed marketplace listing confirms AgentCore integration. | [straiker-12](#fact-straiker-12), [straiker-13](#fact-straiker-13), [straiker-14](#fact-straiker-14) | | Coverage: identity layer | 7 of 10 | 6 | The vendor maps OAuth scopes, flags over-permissioned agents, revokes access and uses ephemeral credentials; vendor claims only. | [straiker-15](#fact-straiker-15), [straiker-16](#fact-straiker-16), [straiker-17](#fact-straiker-17), [straiker-18](#fact-straiker-18) | | Coverage: gateway layer | 3 of 10 | 6 | The vendor integrates with AI gateways but does not document its own gateway or proxy. | [straiker-19](#fact-straiker-19), [straiker-20](#fact-straiker-20), [straiker-21](#fact-straiker-21) | | Stops the action | 7 of 10 | 20 | The vendor blocks exfiltration, code execution and destructive commands and can cut tool access; the third-party report covers taking agents offline, not pre-action blocking. | [straiker-22](#fact-straiker-22), [straiker-23](#fact-straiker-23), [straiker-24](#fact-straiker-24), [straiker-25](#fact-straiker-25) | | Visibility | 7 of 10 | 15 | A real-time inventory of agents, tools, MCP servers and reachable OAuth scopes with audit logs is described; vendor claims only. | [straiker-15](#fact-straiker-15), [straiker-26](#fact-straiker-26), [straiker-27](#fact-straiker-27) | | Deployment | 8 of 10 | 10 | Hook, thin client, extension, API and on-prem options with minutes to deploy are described, and the marketplace listing confirms SaaS delivery. | [straiker-11](#fact-straiker-11), [straiker-14](#fact-straiker-14), [straiker-20](#fact-straiker-20), [straiker-28](#fact-straiker-28), [straiker-29](#fact-straiker-29) | | Policy and governance | 5 of 10 | 10 | Customer-written policies, audit traceability and framework mapping exist; named SIEM, EDR, IdP or ticketing integrations none found. | [straiker-30](#fact-straiker-30), [straiker-31](#fact-straiker-31), [straiker-27](#fact-straiker-27), [straiker-32](#fact-straiker-32) | | Transparency | 3 of 10 | 10 | Research is public; public documentation and a public price none found. | [straiker-33](#fact-straiker-33), [straiker-34](#fact-straiker-34), [straiker-35](#fact-straiker-35) | | Proof | 8 of 10 | 5 | Third parties report $85 million raised and revenue growth, with named customer quotes; Gartner listing is vendor-cited. | [straiker-03](#fact-straiker-03), [straiker-04](#fact-straiker-04), [straiker-36](#fact-straiker-36), [straiker-37](#fact-straiker-37), [straiker-45](#fact-straiker-45) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 2nd | 63.7 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 4th | 65.5 | | [Ranked for the browser layer](/best-for/layer-browser/) | 1st | 65.5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | joint 1st | 68.5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 3rd | 65.5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 14th | 53.5 | ## Product, deployment, price and owner - Product assessed: Straiker platform: Discover AI (inventory and posture), Defend AI (runtime guardrails) and the Agentic Kill Switch; Ascend AI red teaming supports them - Deployment: SaaS, or on premises for regulated environments; hook, API, SDK, webhook, AI sensor, thin client, browser extension or gateway integration - Time to deploy: Minutes, as the vendor says on its product pages - Price: none found on 2026-11-16 on straiker.ai; AWS Marketplace shows one platform dimension under a contract sized with the vendor - Owner: Straiker Inc. (standalone, venture-backed; no parent company found) - Website: straiker.ai ## Facts and sources - **straiker-01** (vendor): Straiker presents itself as "The Agentic Security Company" and says it is staffed from Mountain View and distributed teams.Quote: "Straiker is built by AI security researchers, security engineers, product builders, and enterprise operators working from Mountain View and distributed teams around the world."Source: [straiker.ai](https://www.straiker.ai/about), read 2026-11-16. - **straiker-02** (vendor): Straiker's leadership page lists Ankur Shah as co-founder and CEO and Sreenath Kurupati as co-founder and CTO.Quote: "Ankur Shah is a cybersecurity leader and entrepreneur with a proven track record of building and scaling innovative security solutions."Source: [straiker.ai](https://www.straiker.ai/about), read 2026-11-16. - **straiker-03** (third-party): SiliconANGLE reported on 2026-06-29 that Straiker Inc., based in Mountain View, raised a $64 million Series A led by Marathon Management Partners with Citi Ventures, Illuminate Ventures and Workday Ventures.Quote: "The Series A round was led by Marathon Management Partners alongside Citi Ventures, Illuminate Ventures and Workday Ventures"Source: [siliconangle.com](https://siliconangle.com/2026/06/29/straiker-lands-64m-defend-enterprise-ai-agents-attack/), read 2026-11-16. - **straiker-04** (third-party): On 2026-08-05 Intesa Sanpaolo group venture arm Neva SGR said it joined the $64 million Series A, bringing total capital raised by Straiker to $85 million.Quote: "The transaction brings the total capital raised by the California-based company to"Source: [group.intesasanpaolo.com](https://group.intesasanpaolo.com/en/newsroom/press-releases/2026/08/neva-sgr-invests-in-straiker--a-leader-in-agentic-ai-security), read 2026-11-16. - **straiker-05** (vendor): Straiker's product line is Discover AI, Ascend AI, Defend AI and Straiker Frontiers, plus an Agentic Kill Switch.Quote: "Discover AI maps every agent, tool, MCP, and Skill in your environment."Source: [straiker.ai](https://www.straiker.ai/platform), read 2026-11-16. - **straiker-06** (vendor): Straiker says Discover AI finds sanctioned and shadow AI across endpoint, browser, gateway, cloud and connected SaaS, including coding, productivity and personal agents and local LLMs.Quote: "Find sanctioned and shadow AI across endpoint, browser, gateway, cloud and connected SaaS sources, including coding, productivity and personal agents and local LLMs/runtimes."Source: [straiker.ai](https://www.straiker.ai/products/discover-ai), read 2026-11-16. - **straiker-07** (vendor): Straiker says it supports Cursor Agent, GitHub Copilot, Claude Code and Windsurf through a thin-client integration and detects and blocks exfiltration, remote code execution and destructive infrastructure actions.Quote: "Straiker currently supports Cursor Agent, GitHub Copilot, Claude Code, and Windsurf, with new coding-agent coverage added regularly."Source: [straiker.ai](https://www.straiker.ai/solution/coding-agents), read 2026-11-16. - **straiker-08** (vendor): Straiker says Defend AI blocks actions such as file deletion and configuration changes by coding agents.Quote: "It can detect and block actions such as file deletion and configuration changes while protecting proprietary code and secrets in development environments."Source: [straiker.ai](https://www.straiker.ai/products/defend-ai), read 2026-11-16. - **straiker-09** (vendor): Straiker says its agentic-browser guardrails block prompt injection and govern forms, uploads, downloads, clipboard and screenshots with allowlists and step-level approvals.Quote: "Govern forms, uploads, downloads, clipboard, and screenshots with destination allowlists and step‑level approvals."Source: [straiker.ai](https://www.straiker.ai/solution/runtime-guardrails-for-agentic-web-browsers), read 2026-11-16. - **straiker-10** (vendor): Straiker says Discover AI uses Browser Guard and endpoint or gateway integrations to enforce employee AI usage policies.Quote: "Browser Guard and endpoint/gateway integrations enforce employee usage policies; activity logs capture user sessions, transcripts and supported tool-invocation events."Source: [straiker.ai](https://www.straiker.ai/products/discover-ai), read 2026-11-16. - **straiker-11** (vendor): Straiker says productivity-agent protection deploys through a lightweight sensor, browser extension or API gateway, with most deployments live within minutes.Quote: "Straiker deploys alongside your existing AI productivity agent infrastructure through a lightweight sensor, browser extension, or API gateway."Source: [straiker.ai](https://www.straiker.ai/solution/productivity-agents), read 2026-11-16. - **straiker-12** (vendor): Straiker says it supports Microsoft Copilot (browser, standalone, Office), Copilot Studio, ChatGPT Enterprise and Atlas, Salesforce Agentforce, Slack AI and Claude Cowork and Desktop.Quote: "Straiker currently supports MS Copilot (browser, standalone, Office suite), Copilot Studio, ChatGPT Enterprise and Atlas, Salesforce Agentforce, Slack AI, Claude Cowork + Desktop, and Enterprise Claude"Source: [straiker.ai](https://www.straiker.ai/solution/productivity-agents), read 2026-11-16. - **straiker-13** (vendor): Straiker says it covers custom agents built on AWS Bedrock AgentCore, Azure AI Foundry, Copilot Studio, MCP and LangChain, integrated by API, SDK, webhook or gateway.Quote: "Straiker supports custom AI agents built with AWS Bedrock AgentCore, Azure AI Foundry, Azure Copilot Studio, MCP, LangChain, and custom orchestration frameworks."Source: [straiker.ai](https://www.straiker.ai/solution/custom-built-agents), read 2026-11-16. - **straiker-14** (observed): The AWS Marketplace listing (observed) says Straiker is sold as SaaS and integrates with Amazon Bedrock AgentCore.Quote: "Straiker integrates with Amazon Bedrock AgentCore to help security teams discover agents, continuously test them against real-world attacks, stop malicious or unauthorized behavior at runtime"Source: [aws.amazon.com](https://aws.amazon.com/marketplace/pp/prodview-4ey26i3k5epmi), read 2026-11-16. - **straiker-15** (vendor): Straiker says Discover AI maps the connectors and OAuth scopes productivity agents can reach and helps identify overprivileged access.Quote: "Straiker Discover AI maps the connectors, tools, and data sources that productivity agents access, giving security teams visibility into what each agent can reach and helping identify overprivileged access."Source: [straiker.ai](https://www.straiker.ai/solution/productivity-agents), read 2026-11-16. - **straiker-16** (vendor): Straiker says Discover AI evaluates agent configurations and permissions to surface over-permissioned agents, and the platform blocks or revokes risky MCP and tool access.Quote: "Detect over-permissioned agents, risky MCP connections, and misconfigured integrations across your full agentic ecosystem."Source: [straiker.ai](https://www.straiker.ai/products/discover-ai), read 2026-11-16. - **straiker-17** (vendor): Straiker says its Agentic Kill Switch can revoke an agent's access to tools, data, APIs and systems.Quote: "Revoke access to the tools, data, APIs, and systems the agent can use."Source: [straiker.ai](https://www.straiker.ai/solution/agentic-kill-switch), read 2026-11-16. - **straiker-18** (vendor): Straiker says its agentic-browser guardrails contain cookies, tokens and OAuth scopes per agent run with rotating ephemeral credentials.Quote: "Contain cookies, tokens, and OAuth scopes per agent run, rotating ephemeral credentials to prevent session drift and privilege escalation across tabs or tools."Source: [straiker.ai](https://www.straiker.ai/solution/runtime-guardrails-for-agentic-web-browsers), read 2026-11-16. - **straiker-19** (vendor): Straiker says Defend AI integrates through hooks, APIs and AI gateways.Quote: "Integrations include hooks, APIs and AI gateways."Source: [straiker.ai](https://www.straiker.ai/products/defend-ai), read 2026-11-16. - **straiker-20** (vendor): Straiker says deployment takes minutes with a single hook-based integration or a one-line install via API, SDK, webhook or AI sensor, with no proxies or firewalls required.Quote: "Deploy in minutes with a single hook-based integration. One-line install via API, SDK, webhook, or AI sensor with no thick clients, proxies, firewalls, or infrastructure changes required."Source: [straiker.ai](https://www.straiker.ai/products/defend-ai), read 2026-11-16. - **straiker-21** (vendor): Straiker distinguishes its kill switch from an AI gateway, which it says only controls requests and traffic, and lists gateways among the platforms it protects.Quote: "Protects agents across models, frameworks, clouds, gateways, and architectures"Source: [straiker.ai](https://www.straiker.ai/platform), read 2026-11-16. - **straiker-22** (vendor): Straiker says Defend AI uses detection models to detect or block prompt injection, jailbreaks, data exfiltration and unsafe agent actions at supported execution points.Quote: "Specialized detection models evaluate available conversation and tool context to detect or block prompt injection, jailbreaks, data exfiltration and unsafe agent actions at supported execution points."Source: [straiker.ai](https://www.straiker.ai/products/defend-ai), read 2026-11-16. - **straiker-23** (vendor): Straiker says Defend AI detects remote code execution, file access violations, suspicious outbound access, destructive commands and data exfiltration, in Detect or Protect modes.Quote: "Detects remote code execution, file access violations, suspicious outbound access, destructive commands, and data exfiltration"Source: [straiker.ai](https://www.straiker.ai/platform), read 2026-11-16. - **straiker-24** (vendor): Straiker says its Agentic Kill Switch cuts off tool access, suspends the session, isolates or takes an agent offline, and preserves the audit trail.Quote: "Cut off tool access, suspend the session, isolate the agent, or shut it down. Stop one agent or all of them, with a record of what happened."Source: [straiker.ai](https://www.straiker.ai/solution/agentic-kill-switch), read 2026-11-16. - **straiker-25** (third-party): The European reported that the Agentic Kill Switch, launched in August 2026, can take a compromised agent offline in seconds.Quote: "Straiker says the kill switch can take it offline in seconds and return control to human operators."Source: [the-european.eu](https://the-european.eu/story-64052/ai-kill-switch-launched-to-stop-rogue-agents-in-seconds.html), read 2026-11-16. - **straiker-26** (vendor): Straiker says Discover AI automatically catalogs agents, tools, MCP servers and integrations in a real-time inventory.Quote: "Automatically identify and catalog every AI agent operating across your environment. AI agent discovery builds a real-time inventory of agents, tools, MCP servers, and integrations."Source: [straiker.ai](https://www.straiker.ai/products/discover-ai), read 2026-11-16. - **straiker-27** (vendor): Straiker says Defend AI provides dashboards, audit logs and alerts over Slack, email or webhook.Quote: "Dashboards, audit logs, and instant alerts over Slack, email, or webhook keep teams informed and audit-ready."Source: [straiker.ai](https://www.straiker.ai/products/defend-ai), read 2026-11-16. - **straiker-28** (vendor): Straiker says coding-agent protection uses a thin client that deploys in minutes next to IDE and terminal workflows.Quote: "Straiker uses a thin-client integration that deploys in minutes and runs alongside existing IDE and terminal workflows."Source: [straiker.ai](https://www.straiker.ai/solution/coding-agents), read 2026-11-16. - **straiker-29** (vendor): Straiker says Defend AI is offered as SaaS or on premises for regulated and sovereign environments.Quote: "SaaS or on premises for regulated and sovereign environments"Source: [straiker.ai](https://www.straiker.ai/platform), read 2026-11-16. - **straiker-30** (vendor): Straiker says Defend AI offers Detect or Protect modes with policies customers write themselves.Quote: "Detect or Protect modes, with policies you write yourself"Source: [straiker.ai](https://www.straiker.ai/platform), read 2026-11-16. - **straiker-31** (vendor): Straiker says it produces audit-grade traceability across AI interactions and maps to the EU AI Act, NIST AI 600-1, PCI-DSS v4.0, HIPAA and MITRE ATLAS.Quote: "Straiker produces audit-grade traceability across every AI interaction, decision, and action"Source: [straiker.ai](https://www.straiker.ai/solution/ai-compliance-governance), read 2026-11-16. - **straiker-32** (none-found): None found on 2026-11-16: named SIEM, EDR, identity provider or ticketing integrations on the partner page; only AI platforms, gateways, clouds and Slack, email or webhook alerts were listed.Source: [straiker.ai](https://www.straiker.ai/partner/technology-and-channel-partners), read 2026-11-16. - **straiker-33** (none-found): None found on 2026-11-16: public product documentation; docs.straiker.ai loads an access-code login page.Source: [docs.straiker.ai](https://docs.straiker.ai/), read 2026-11-16. - **straiker-34** (observed): The AWS Marketplace listing (observed) shows a single "Straiker Platform" pricing dimension billed under a 12-month contract sized with the vendor, not a public price.Quote: "This listing has one pricing dimension: Straiker Platform, billed by units of platform access under a contract."Source: [aws.amazon.com](https://aws.amazon.com/marketplace/pp/prodview-4ey26i3k5epmi), read 2026-11-16. - **straiker-35** (vendor): Straiker publishes research on its site, including a Threat Research report and blog posts dated through 2026-10-21, readable without sign-in.Quote: "Explore Straiker STAR’s latest findings on LLM vulnerabilities, agentic AI threats, and runtime guardrails."Source: [straiker.ai](https://www.straiker.ai/research), read 2026-11-16. - **straiker-36** (vendor): Straiker's customer page quotes security leaders at Omada Health, Coupa, American Express Global Business Travel, EnterpriseDB and Automation Anywhere.Quote: "Straiker helped remove the security blindspot in our B2B AI Application with their unique approach with its Ascend AI product."Source: [straiker.ai](https://www.straiker.ai/customers), read 2026-11-16. - **straiker-37** (vendor): Straiker says it was listed as a Sample Vendor in Agentic AI Security in two Gartner Hype Cycle reports published 2026-10-12.Quote: "Straiker is listed as a Sample Vendor in the Agentic AI Security category of two Gartner Hype Cycle reports published October 12, 2026"Source: [straiker.ai](https://www.straiker.ai/products/defend-ai), read 2026-11-16. - **straiker-38** (vendor): Straiker lists awards including Best Cybersecurity Startup and Best AI Security Testing Platform at The Hacker News Cybersecurity Stars Awards 2026 and the CB Insights AI 100 2026.Quote: "At The Hacker News Cybersecurity Stars Awards 2026, Straiker took two of the program's top honors: Best Cybersecurity Startup and Best AI Security Testing Platform for Ascend AI."Source: [straiker.ai](https://www.straiker.ai/awards-and-recognition), read 2026-11-16. - **straiker-45** (third-party): On 2026-06-29 SiliconANGLE reported Straiker says run-rate revenue grew more than 15-fold in under a year and that it counts frontier AI labs and Fortune 500 companies as customers.Quote: "It counts frontier AI labs and Fortune 500 companies among its customers."Source: [siliconangle.com](https://siliconangle.com/2026/06/29/straiker-lands-64m-defend-enterprise-ai-agents-attack/), read 2026-11-16. - **straiker-46** (vendor): On 2026-08-11 Straiker announced it is an OpenAI Select Partner, and its partner page says it is an Anthropic Compliance Partner integrated with the Claude Compliance API.Quote: "Straiker is an Anthropic Compliance Partner, integrated with the Claude Compliance API."Source: [straiker.ai](https://www.straiker.ai/partner/technology-and-channel-partners), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [SentinelOne (Prompt Security) profile](/vendors/sentinelone/) - [WitnessAI profile](/vendors/witnessai/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # WitnessAI AI agent security URL: https://aiagentsecuritysolutions.com/vendors/witnessai/ Updated: 2026-11-16 WitnessAI enforces at the network rather than on the device. It says a tool or MCP server not on the approved list is denied before the tool executes, and every blocked call records the user, agent, tool and rule in an audit trail. A third-party report confirms the network enforcement and the approved list, and it captures Cursor, Claude Code and Codex activity by network analysis without an endpoint client. Deployment is agentless, with F5 BIG-IP SSL Orchestrator integration and REST APIs. AI browser coverage, named SaaS agent platforms, agent credential revocation, SIEM integrations, public documentation and a price: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Denies off-list MCP servers and tools before they execute at the network, with an audit record of user, agent, tool and rule per blocked call (witnessai-14, witnessai-40). **Weakness:** None found on 2026-11-16 for AI browser coverage, named SaaS agent platforms, agent credential inventory or revocation, SIEM integrations, public documentation or a price (witnessai-27, witnessai-30, witnessai-32, witnessai-19, witnessai-41). ## Scores and evidence WitnessAI is 15th in the overall ranking at 48.9 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **WitnessAI: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 5 of 10 | 6 | The vendor captures Cursor, Claude Code and Codex activity via network analysis, without an endpoint client; local enforcement detail is limited. | [witnessai-25](#fact-witnessai-25), [witnessai-26](#fact-witnessai-26) | | Coverage: browser layer | 2 of 10 | 6 | Only AI extensions and plugins are mentioned; browser agent or AI browser coverage none found. | [witnessai-27](#fact-witnessai-27), [witnessai-28](#fact-witnessai-28) | | Coverage: cloud workspace layer | 2 of 10 | 6 | Custom public-cloud agents are discovered; named SaaS agent platform coverage none found. | [witnessai-29](#fact-witnessai-29), [witnessai-30](#fact-witnessai-30) | | Coverage: identity layer | 2 of 10 | 6 | Agent activity is attributed to humans; credential inventory, least privilege and revocation none found. | [witnessai-31](#fact-witnessai-31), [witnessai-32](#fact-witnessai-32) | | Coverage: gateway layer | 8 of 10 | 6 | Network enforcement of an approved MCP server and tool list plus REST APIs is described, and a third-party report confirms it. | [witnessai-14](#fact-witnessai-14), [witnessai-33](#fact-witnessai-33), [witnessai-34](#fact-witnessai-34), [witnessai-36](#fact-witnessai-36) | | Stops the action | 7 of 10 | 20 | Unapproved tool and MCP calls are denied before execution and prompt injection is blocked; action types beyond tool access are limited. | [witnessai-14](#fact-witnessai-14), [witnessai-35](#fact-witnessai-35), [witnessai-36](#fact-witnessai-36) | | Visibility | 6 of 10 | 15 | The vendor shows MCP servers, tools exposed and every tool called; per-agent permissions and data reach are limited. | [witnessai-08](#fact-witnessai-08), [witnessai-16](#fact-witnessai-16), [witnessai-26](#fact-witnessai-26), [witnessai-37](#fact-witnessai-37) | | Deployment | 5 of 10 | 10 | Network-level agentless deployment with F5 integration is described; time to deploy and requirements none found. | [witnessai-06](#fact-witnessai-06), [witnessai-09](#fact-witnessai-09), [witnessai-38](#fact-witnessai-38), [witnessai-39](#fact-witnessai-39) | | Policy and governance | 4 of 10 | 10 | One approved list and an audit trail of blocked calls exist; SIEM, EDR, IdP and ticketing integrations none found. | [witnessai-34](#fact-witnessai-34), [witnessai-40](#fact-witnessai-40), [witnessai-41](#fact-witnessai-41) | | Transparency | 2 of 10 | 10 | A survey report is published; public documentation and a price none found. | [witnessai-19](#fact-witnessai-19), [witnessai-20](#fact-witnessai-20), [witnessai-21](#fact-witnessai-21) | | Proof | 7 of 10 | 5 | A third party reports $85 million raised, with a named case study, an IDC Innovators excerpt and a ServiceNow alliance cited by the vendor. | [witnessai-03](#fact-witnessai-03), [witnessai-22](#fact-witnessai-22), [witnessai-23](#fact-witnessai-23), [witnessai-24](#fact-witnessai-24) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | 15th | 48.9 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | 15th | 52.5 | | [Ranked for the browser layer](/best-for/layer-browser/) | joint 14th | 43.5 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | 16th | 43.5 | | [Ranked for the identity layer](/best-for/layer-identity/) | 15th | 43.5 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 8th | 61.5 | ## Product, deployment, price and owner - Product assessed: WitnessAI Agentic Control (Observe, Control and Protect for agents, MCP Catalog, runtime protection REST APIs) - Deployment: Network-level, agentless discovery and enforcement (no endpoint client or browser extension), F5 BIG-IP SSL Orchestrator integration, and REST APIs for custom agents - Time to deploy: none found on 2026-11-16 - Price: none found on 2026-11-16 (pricing page returns 404; demo request only) - Owner: WitnessAI (standalone, venture-backed; no parent company stated on its pages) - Website: [witness.ai](https://witness.ai) ## Facts and sources - **witnessai-01** (vendor): WitnessAI describes itself as an AI security and governance platform for employees, models, applications and agents, sold as Observe, Control and Protect (with Attack for red teaming).Quote: "WitnessAI makes AI safe to use and deploy for every employee, model, application, and agent."Source: [witness.ai](https://witness.ai/), read 2026-11-16. - **witnessai-02** (vendor): WitnessAI lists Rick Caccia as co-founder and CEO on its own leadership page.Quote: "Chief Executive Officer & Co-founder"Source: [witness.ai](https://witness.ai/about-us/), read 2026-11-16. - **witnessai-03** (third-party): SecurityWeek reported on 2026-01-13 that WitnessAI raised $58 million, bringing total funding to over $85 million, with a standalone investor round led by Sound Ventures.Quote: "announced raising $58 million in strategic funding, which brings the total raised by the company to over $85 million."Source: [securityweek.com](https://www.securityweek.com/witnessai-raises-58-million-for-ai-security-platform/), read 2026-11-16. - **witnessai-06** (vendor): WitnessAI says it sees AI activity at the network level without browser extensions or endpoint clients, including native apps such as Windows Copilot and Office 365.Quote: "See AI activity across your entire network without relying on browser extensions or endpoint clients."Source: [witness.ai](https://witness.ai/observe/), read 2026-11-16. - **witnessai-08** (vendor): WitnessAI says it shows which agentic plugins employees install and which MCP servers they reach.Quote: "See which agentic plugins employees install and which MCP servers they reach."Source: [witness.ai](https://witness.ai/for-employees/), read 2026-11-16. - **witnessai-09** (vendor): WitnessAI says it works over network connectivity and has a named integration with F5 BIG-IP SSL Orchestrator to see encrypted AI traffic.Quote: "WitnessAI uses network connectivity to observe, control, and protect employees' use of AI applications."Source: [witness.ai](https://witness.ai/resources/witnessai-announces-integration-with-f5-big-ip-to-deliver-visibility-and-control-over-encrypted-genai-activity/), read 2026-11-16. - **witnessai-14** (vendor): WitnessAI says a tool or MCP server not on the approved list is denied before the tool executes, whichever model provider the call is routed through.Quote: "Anything off the list is denied before the tool executes"Source: [witness.ai](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/), read 2026-11-16. - **witnessai-16** (vendor): On 2026-06-17 WitnessAI announced an MCP Catalog that scores known tools against OWASP and CVE risk classes so teams can assess risk before approval.Quote: "WitnessAI's new MCP Catalog scores known tools against OWASP and CVE risk classes, enabling security teams to assess risk before granting approval."Source: [prnewswire.com](https://www.prnewswire.com/news-releases/witnessai-introduces-agentic-control-to-secure-and-govern-ai-agents-and-mcp-servers-302802606.html), read 2026-11-16. - **witnessai-17** (none-found): None found on 2026-11-16: a per-application risk rating for web AI apps (data use, training on inputs, vendor posture) or an exportable inventory report, on the Observe and Product pages.Source: [witness.ai](https://witness.ai/observe/), read 2026-11-16. - **witnessai-19** (none-found): None found on 2026-11-16: public documentation: docs.witness.ai loaded as a password-gated page ("Please enter your password for this site") with no readable documentation.Source: [docs.witness.ai](https://docs.witness.ai/), read 2026-11-16. - **witnessai-20** (none-found): None found on 2026-11-16: a published price: https://witness.ai/pricing/ returned a 404 page, and no price appeared on the Product, Observe, Control or Protect pages; the pages direct buyers to book a demo.Source: [witness.ai](https://witness.ai/pricing/), read 2026-11-16. - **witnessai-21** (vendor): On 2026-07-22 WitnessAI published a survey report, The Hidden Cost of Enterprise AI, finding that few enterprises formally inventory and approve all agents.Quote: "fewer than one in five (18%) report that all agents are formally inventoried and approved by their security team"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/new-witnessai-report-reveals-the-financial-cost-and-risk-factors-of-enterprise-ai-adoption-302831518.html), read 2026-11-16. - **witnessai-22** (vendor): WitnessAI publishes a customer case study with InComm Payments, with its CISO quoted on compliance, data-loss prevention and privacy visibility.Quote: "Our compliance, data-loss prevention, and privacy teams now have total visibility and confidence in our AI security."Source: [witness.ai](https://witness.ai/resources/incomm-payments/), read 2026-11-16. - **witnessai-23** (vendor): WitnessAI says it was named a founding member of the ServiceNow Control Tower Alliance.Quote: "WitnessAI Named a Founding Member of ServiceNow Control Tower Alliance"Source: [witness.ai](https://witness.ai/for-developers/), read 2026-11-16. - **witnessai-24** (vendor): WitnessAI hosts an excerpt of IDC Innovators: Security for Agentic AI, 2025, an analyst document covering emerging vendors in this area.Quote: "This IDC Innovators presentation highlights emerging vendors addressing the security challenges of agentic AI"Source: [witness.ai](https://witness.ai/resources/idc-innovators-security-for-agentic-ai-2025/), read 2026-11-16. - **witnessai-25** (vendor): WitnessAI says it captures prompts, tool calls and shell executions across supported Cursor, Claude Code and Codex workflows.Quote: "Capture prompts, tool calls, and shell executions across supported Cursor, Claude Code, and Codex workflows."Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-26** (vendor): WitnessAI says it discovers agentic activity in Claude Desktop, VS Code with AI extensions, ChatGPT plugins and local agent frameworks on developer machines, via network analysis rather than an endpoint client.Quote: "local agents running on developer machines (including LangChain, LlamaIndex, CrewAI, AutoGPT, and custom implementations)."Source: [witness.ai](https://witness.ai/blog/introducing-witnessai-agentic-security-extending-the-confidence-layer-to-ai-agents/), read 2026-11-16. - **witnessai-27** (none-found): None found on 2026-11-16: coverage of browser agents or AI browsers on the For Agents, Observe, Control and Product pages; the pages say the platform works without browser extensions.Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-28** (vendor): WitnessAI says it discovers agentic activity in VS Code with AI extensions and ChatGPT with enabled plugins; this is about AI extensions and plugins, not a described AI-browser product.Quote: "VSCode with AI extensions, ChatGPT with enabled plugins, and local agents running on developer machines"Source: [witness.ai](https://witness.ai/blog/introducing-witnessai-agentic-security-extending-the-confidence-layer-to-ai-agents/), read 2026-11-16. - **witnessai-29** (vendor): On 2026-06-17 WitnessAI said it discovers agents including custom agents built in the public cloud and the MCP servers they reach.Quote: "custom agents built in the public cloud, and identifies the specific MCP servers, tools, and downstream systems each agent reaches."Source: [prnewswire.com](https://www.prnewswire.com/news-releases/witnessai-introduces-agentic-control-to-secure-and-govern-ai-agents-and-mcp-servers-302802606.html), read 2026-11-16. - **witnessai-30** (none-found): None found on 2026-11-16: named coverage of Microsoft Copilot Studio, Salesforce Agentforce, ServiceNow or AWS Bedrock agents on the For Agents and Product pages.Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-31** (vendor): WitnessAI says it connects human and agent identities and attributes agent-to-agent activity to the human who started the workflow.Quote: "It connects human and agent identities and captures decision-making context at runtime, including agent state and execution commands."Source: [witness.ai](https://witness.ai/blog/introducing-witnessai-agentic-security-extending-the-confidence-layer-to-ai-agents/), read 2026-11-16. - **witnessai-32** (none-found): None found on 2026-11-16: an inventory of agent credentials or tokens, least-privilege review, or credential revocation on the For Agents, Control and Product pages.Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-33** (vendor): WitnessAI says custom agents get runtime protection through lightweight REST APIs added to the orchestration layer.Quote: "Add runtime protection to agent orchestration through lightweight REST APIs."Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-34** (vendor): WitnessAI says one organization-wide approved list of MCP servers and tools is enforced for agent traffic that traverses the platform.Quote: "WitnessAI enforces it for agent traffic that traverses the platform, permitting agentic work while denying specific tools and servers."Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-35** (vendor): WitnessAI says runtime protection blocks prompt injection and jailbreak attempts before agents act.Quote: "Runtime protection blocks prompt injection and jailbreak attempts before agents act."Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-36** (third-party): Help Net Security reported on 2026-06-17 that WitnessAI Agentic Control gives visibility and real-time network enforcement over the tools and MCP servers agents can access.Quote: "deep visibility and real-time network enforcement over the tools and MCP servers AI agents can access"Source: [helpnetsecurity.com](https://www.helpnetsecurity.com/2026/06/17/witnessai-agentic-control/), read 2026-11-16. - **witnessai-37** (vendor): WitnessAI says it shows the public and private MCP servers agents reach, the tools each exposes and every tool called.Quote: "See the public and private MCP servers agents reach, the tools each exposes, and every tool called as agents work."Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-38** (vendor): WitnessAI says it discovers and governs supported agent traffic at the network, and custom agents attach protection in the orchestration layer.Quote: "WitnessAI discovers and governs supported agent traffic at the network. Custom agents receive bidirectional runtime protection in the orchestration layer."Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-39** (none-found): None found on 2026-11-16: a published time to deploy, system requirements, or supported operating systems on the Product, For Agents and Control pages.Source: [witness.ai](https://witness.ai/product/), read 2026-11-16. - **witnessai-40** (vendor): WitnessAI says every blocked tool call records the user, agent, tool and rule in one policy framework and audit trail.Quote: "Every blocked tool call records the user, agent, tool, and rule."Source: [witness.ai](https://witness.ai/for-agents/), read 2026-11-16. - **witnessai-41** (none-found): None found on 2026-11-16: named SIEM, EDR, identity provider or ticketing integrations on the Control, For Agents and Product pages (an F5 BIG-IP integration and a ServiceNow alliance are named elsewhere).Source: [witness.ai](https://witness.ai/control/), read 2026-11-16. - **witnessai-42** (vendor): On 2026-06-17 WitnessAI said Agentic Control was available to all customers and cited a 99.3% true positive rate on its own guardrail benchmark.Quote: "The WitnessAI platform delivers a 99.3% true positive rate on employee AI guardrails"Source: [prnewswire.com](https://www.prnewswire.com/news-releases/witnessai-introduces-agentic-control-to-secure-and-govern-ai-agents-and-mcp-servers-302802606.html), read 2026-11-16. ## Related pages - [All vendor profiles](/vendors/) - [Straiker profile](/vendors/straiker/) - [Zenity profile](/vendors/zenity/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # Zenity AI agent security URL: https://aiagentsecuritysolutions.com/vendors/zenity/ Updated: 2026-11-16 Zenity lists a broad set of platforms it covers, including Bedrock AgentCore, ChatGPT Enterprise, Claude Enterprise and Microsoft Copilot, under one runtime Boundaries engine that evaluates each agent action and can let it through, modify it, block it or shut the agent down. Agent identities are pulled from Entra ID and Okta with their privilege level and used to gate actions, and static keys are flagged. Coding agents connect by native hooks and OpenTelemetry, and an MCP gateway uses one registered URL. Documented browser control, readable public documentation, a price and a time to deploy: none found on 16 Nov 2026. ## Strength and weakness for AI Agent Security Solutions **Strength:** Covers agent identity, SaaS and cloud agent platforms, coding agents and an MCP gateway under one runtime Boundaries engine that can block, modify or shut down an agent (zenity-45, zenity-34, zenity-30, zenity-24). **Weakness:** Public docs sit behind a login, no price or time to deploy is published, and browser-agent control is none found on 2026-11-16 (zenity-39, zenity-59, zenity-40). ## Scores and evidence Zenity is joint 8th in the overall ranking at 58.4 of 100. Each score below lists the fact ids behind it; the facts follow further down with their sources and read dates. **Zenity: score per criterion** | Criterion | Score | Weight | Why | Facts | |---|---|---|---|---| | Coverage: endpoint layer | 7 of 10 | 6 | Named coding agents are discovered and dangerous tool calls blocked via native hooks; OS and install detail none found. | [zenity-19](#fact-zenity-19), [zenity-20](#fact-zenity-20), [zenity-26](#fact-zenity-26), [zenity-50](#fact-zenity-50) | | Coverage: browser layer | 1 of 10 | 6 | Only research on agentic browsers; documented browser control none found. | [zenity-40](#fact-zenity-40), [zenity-41](#fact-zenity-41) | | Coverage: cloud workspace layer | 7 of 10 | 6 | A broad list of SaaS and cloud agent platforms is named with one rule across them; third-party coverage concerns research, not coverage. | [zenity-30](#fact-zenity-30), [zenity-31](#fact-zenity-31), [zenity-47](#fact-zenity-47), [zenity-12](#fact-zenity-12) | | Coverage: identity layer | 7 of 10 | 6 | Agent identities are pulled from Entra and Okta with privilege level, used to gate actions, and static keys are flagged; vendor claims only. | [zenity-34](#fact-zenity-34), [zenity-35](#fact-zenity-35), [zenity-36](#fact-zenity-36) | | Coverage: gateway layer | 7 of 10 | 6 | An MCP gateway can allow, modify or block tool calls via one registered URL; vendor claims only. | [zenity-24](#fact-zenity-24), [zenity-25](#fact-zenity-25), [zenity-48](#fact-zenity-48) | | Stops the action | 7 of 10 | 20 | The vendor blocks exfiltration chains, privilege escalation and dangerous tool calls before execution; vendor claims only. | [zenity-45](#fact-zenity-45), [zenity-46](#fact-zenity-46), [zenity-50](#fact-zenity-50), [zenity-52](#fact-zenity-52) | | Visibility | 7 of 10 | 15 | A live agent inventory with data access and per-step activity tables is described; vendor claims only. | [zenity-18](#fact-zenity-18), [zenity-27](#fact-zenity-27), [zenity-43](#fact-zenity-43), [zenity-54](#fact-zenity-54) | | Deployment | 5 of 10 | 10 | API connection, native hooks and OpenTelemetry, and a single gateway URL are described with inventory on connect; OS list and time to deploy none found. | [zenity-25](#fact-zenity-25), [zenity-26](#fact-zenity-26), [zenity-42](#fact-zenity-42), [zenity-29](#fact-zenity-29), [zenity-58](#fact-zenity-58) | | Policy and governance | 5 of 10 | 10 | Identity-gated rules, preview mode and audit trails exist; documented SIEM and ticketing export none found. | [zenity-35](#fact-zenity-35), [zenity-37](#fact-zenity-37), [zenity-49](#fact-zenity-49), [zenity-28](#fact-zenity-28), [zenity-60](#fact-zenity-60) | | Transparency | 3 of 10 | 10 | Research is published and covered by third parties; readable documentation and a price none found. | [zenity-39](#fact-zenity-39), [zenity-59](#fact-zenity-59), [zenity-12](#fact-zenity-12), [zenity-14](#fact-zenity-14) | | Proof | 7 of 10 | 5 | A third party reports a $125 million round, SoftBank is named and analyst recognition is vendor-cited. | [zenity-05](#fact-zenity-05), [zenity-04](#fact-zenity-04), [zenity-08](#fact-zenity-08) | ## Where it ranks | View | Rank | Total | |---|---|---| | [Overall ranking](/ranking/) | joint 8th | 58.4 | | [Ranked for the endpoint layer](/best-for/layer-endpoint/) | joint 10th | 62.0 | | [Ranked for the browser layer](/best-for/layer-browser/) | 13th | 44.0 | | [Ranked for the cloud workspace layer](/best-for/layer-cloud/) | joint 5th | 62.0 | | [Ranked for the identity layer](/best-for/layer-identity/) | 6th | 62.0 | | [Ranked for the gateway layer](/best-for/layer-gateway/) | 7th | 62.0 | ## Product, deployment, price and owner - Product assessed: Zenity AI Agent Security and Governance Platform: AI Observability, AISPM, AI Exposure Management, Boundaries (runtime enforcement), Agentic Identity, MCP Security and AIDR - Deployment: Connects to platforms by API and integration (Microsoft, Salesforce, OpenAI, Google, AWS, Okta and others per docs navigation); coding agents via native hooks, OpenTelemetry and the Zenity MCP gateway; no agent or proxy install details published - Time to deploy: none found on 2026-11-16 - Price: none found on 2026-11-16 - Owner: Zenity (standalone, venture-backed company; no parent company stated on its pages) - Website: [zenity.io](https://zenity.io) ## Facts and sources - **zenity-01** (vendor): Zenity describes itself as an AI agent security platform that spans discovery, policy, identity, and runtime defense.Quote: "Zenity secures long-horizon AI agents at the decision, the exact point where context and intent come together to create the highest risk."Source: [zenity.io](https://zenity.io/), read 2026-11-16. - **zenity-02** (vendor): On 2026-08-03, Zenity announced a $125 million Series C led by Norwest; it is a standalone venture-backed company headquartered in New York.Quote: "today announced a $125 million Series C led by Norwest."Source: [zenity.io](https://zenity.io/press-release/zenity-raises-125-million-to-secure-the-era-of-1-billion-ai-agents), read 2026-11-16. - **zenity-03** (vendor): Zenity lists its funding history on its company page: a $5M seed in 2021, a $16.5M Series A in 2023 and a $38M Series B in 2024.Quote: "Zenity raises $38M Series B round of fundraising, co-led by Third Point Ventures and DTCP"Source: [zenity.io](https://zenity.io/company), read 2026-11-16. - **zenity-04** (vendor): Zenity says most of its customers are Fortune 500 and Global 2000 organizations and names SoftBank Corp. as a customer.Quote: "including SoftBank Corp."Source: [zenity.io](https://zenity.io/press-release/zenity-raises-125-million-to-secure-the-era-of-1-billion-ai-agents), read 2026-11-16. - **zenity-05** (third-party): Israeli business outlet Calcalist CTech, on 2026-09-03, reports that Zenity recently raised $125 million in a round led by Norwest.Quote: "a company that recently raised $125 million in a funding round led by Norwest"Source: [calcalistech.com](https://www.calcalistech.com/ctechnews/article/rsdllsuy3), read 2026-11-16. - **zenity-06** (vendor): On 2026-09-15, Zenity announced recognition in two independent analyst reports, from Latio and Software Analyst Cyber Research (SACR).Quote: "today announced recognition in two independent analyst reports examining the evolving AI agent security market."Source: [zenity.io](https://zenity.io/press-release/zenity-earns-recognition-in-ai-agent-security-reports-from-latio-and-sacr), read 2026-11-16. - **zenity-07** (vendor): Zenity says SACR places it among vendors combining runtime behavior analysis, MCP and tool governance, agent identity, audit evidence and response.Quote: "SACR identifies Zenity among vendors that combine multiple agent security capabilities, including runtime behavior analysis, MCP and tool governance"Source: [zenity.io](https://zenity.io/press-release/zenity-earns-recognition-in-ai-agent-security-reports-from-latio-and-sacr), read 2026-11-16. - **zenity-08** (vendor): On 2026-09-23, Zenity announced it was named a Market Shaper in Gartner's first Emerging Market Quadrant for AI Application Security (startup vendors).Quote: "Zenity has been named a Market Shaper in Gartner"Source: [zenity.io](https://zenity.io/company/newsroom/press-releases), read 2026-11-16. - **zenity-09** (vendor): Zenity lists a Gartner paper naming it the "Company to Beat" in AI agent governance on its platform pages.Quote: "Gartner® Names Zenity as the Company to Beat in AI Agent Governance"Source: [zenity.io](https://zenity.io/platform/mcp-security), read 2026-11-16. - **zenity-10** (vendor): On 2026-01-06, Zenity announced it is available on AWS Marketplace with support for Amazon Bedrock AgentCore, and says its coverage spans cloud, SaaS, endpoints and agentic browsers.Quote: "across cloud, SaaS, endpoints, and agentic browsers"Source: [zenity.io](https://zenity.io/press-release/zenity-now-available-on-aws-marketplace), read 2026-11-16. - **zenity-11** (vendor): On 2025-03-11, Zenity announced it is available in the Microsoft Azure Marketplace.Quote: "now available in the Microsoft Azure Marketplace"Source: [zenity.io](https://zenity.io/press-release/zenity-now-available-in-the-microsoft-azure-marketplace), read 2026-11-16. - **zenity-12** (third-party): On 2026-09-24, The Register reported that Zenity Labs found three Salesforce Agentforce flaws (SalesBleed) and reported them to Salesforce, which fixed them.Quote: "Zenity Labs uncovered the three vulnerabilities, collectively called SalesBleed, and reported them to Salesforce"Source: [theregister.com](https://www.theregister.com/security/2026/09/24/salesforce-agentforce-vulns-allowed-0-click-crm-data-theft-anonymous-phishing/5298958), read 2026-11-16. - **zenity-13** (third-party): On 2026-09-25, SecurityWeek covered the SalesBleed Agentforce flaws that Zenity Labs reported.Quote: "Three vulnerabilities in Salesforce Agentforce could have allowed attackers to hijack trusted agents for sensitive CRM data exfiltration and phishing, Zenity Labs reports."Source: [securityweek.com](https://www.securityweek.com/salesbleed-flaws-in-salesforce-agentforce-enabled-zero-click-data-exfiltration/), read 2026-11-16. - **zenity-14** (vendor): On 2026-10-08, Zenity Labs published research on AgentCorruption, a chain of AWS AgentCore flaws.Quote: "Zenity Labs Discloses AgentCorruption, a Chain of AWS AgentCore Flaws That Allowed One Prompt to Take Over All AgentCore Agents"Source: [zenity.io](https://zenity.io/company/newsroom/press-releases), read 2026-11-16. - **zenity-18** (vendor): Zenity says its MCP Security product imports servers from its inventory or public registries so known and shadow servers show in one place.Quote: "Import and discovery: Add servers from your Zenity inventory or from public registries, so both known and shadow servers surface in one place."Source: [zenity.io](https://zenity.io/platform/mcp-security), read 2026-11-16. - **zenity-19** (vendor): Zenity says it discovers coding and personal agents on developer laptops and desktops, plus the MCP servers, tools and data they connect to.Quote: "Discover every coding and personal agent across developer laptops and desktops, and the MCP servers, tools, and data they connect to."Source: [zenity.io](https://zenity.io/use-cases/agent-type/coding-personal-agents), read 2026-11-16. - **zenity-20** (vendor): Zenity names Claude Code and Cowork, ChatGPT Codex, GitHub Copilot and Cursor as the coding and personal agents where it enforces boundaries.Quote: "enforces hard boundaries across Anthropic's Claude Code and Cowork, ChatGPT Codex, GitHub Copilot, and Cursor."Source: [zenity.io](https://zenity.io/use-cases/agent-type/coding-personal-agents), read 2026-11-16. - **zenity-21** (vendor): Zenity says AI Observability extends visibility to coding assistants and local MCPs running outside centrally managed platforms.Quote: "AI Observability extends visibility to coding assistants and local MCPs running outside centrally managed platforms."Source: [zenity.io](https://zenity.io/platform/ai-observability), read 2026-11-16. - **zenity-24** (vendor): Zenity says its MCP product can allow, modify or block a tool call based on live context.Quote: "Allow, modify, or block a tool call based on live context, not a static allow-list."Source: [zenity.io](https://zenity.io/platform/mcp-security), read 2026-11-16. - **zenity-25** (vendor): Zenity says MCP traffic is controlled by publishing a server and registering one gateway URL, so agents keep working unchanged.Quote: "Publish a server, register one gateway URL, and agents keep working unchanged."Source: [zenity.io](https://zenity.io/platform/mcp-security), read 2026-11-16. - **zenity-26** (vendor): Zenity says it collects coding-agent telemetry through native hooks and OpenTelemetry.Quote: "through native hooks and OpenTelemetry: each step an agent takes, the code and data it touches"Source: [zenity.io](https://zenity.io/use-cases/agent-type/coding-personal-agents), read 2026-11-16. - **zenity-27** (vendor): Zenity says its MCP Live View shows each request, tool invoked and identity behind it in real time.Quote: "Live View shows each request, the tool invoked, and the identity behind it, in real time."Source: [zenity.io](https://zenity.io/platform/mcp-security), read 2026-11-16. - **zenity-28** (vendor): Zenity says it provides unified visibility, audit trails and enforcement across coding and personal agents.Quote: "Prove governance across every coding and personal agent, with unified visibility, audit trails, and enforcement."Source: [zenity.io](https://zenity.io/use-cases/agent-type/coding-personal-agents), read 2026-11-16. - **zenity-29** (none-found): None found on 2026-11-16: operating systems supported, endpoint install method or time to deploy on developer machines, searched on the coding and personal agents page, MCP security page and platform page.Source: [zenity.io](https://zenity.io/use-cases/agent-type/coding-personal-agents), read 2026-11-16. - **zenity-30** (vendor): Zenity lists the platforms it covers, including Bedrock AgentCore, ChatGPT Enterprise, Claude Enterprise, Copilot Studio, Vertex AI, Microsoft 365 Copilot, Foundry, Power Platform, Salesforce Agentforce and ServiceNow.Quote: "Amazon Bedrock AgentCore ChatGPT Enterprise Claude Enterprise Copilot Studio Google Vertex AI Microsoft 365 Copilot Microsoft Foundry Power Platform Salesforce Salesforce Agentforce ServiceNow"Source: [zenity.io](https://zenity.io/), read 2026-11-16. - **zenity-31** (vendor): Zenity says it secures agents inside SaaS platforms such as Microsoft 365 Copilot, Agentforce, Copilot Studio and ChatGPT Enterprise.Quote: "Zenity secures the agents living inside these platforms: what they were built to do, what they can access, and what they actually do once they're running"Source: [zenity.io](https://zenity.io/use-cases/agent-type/agentic-saas), read 2026-11-16. - **zenity-32** (vendor): Zenity says AI Observability inventories agents deployed across cloud platforms and frameworks.Quote: "AI Observability inventories every agent deployed across an organization's cloud platforms and frameworks."Source: [zenity.io](https://zenity.io/use-cases/agent-type/cloud-and-homegrown), read 2026-11-16. - **zenity-33** (vendor): On 2026-06-12, Zenity announced an integration with the Claude Compliance API for Claude Enterprise.Quote: "integration with Claude's Compliance API that extends governance and security controls for organizations using Claude Enterprise"Source: [zenity.io](https://zenity.io/press-release/zenity-extends-ai-agent-security-and-governance-to-claude-enterprise), read 2026-11-16. - **zenity-34** (vendor): Zenity says it pulls each agent's assumed identities from Entra ID and Okta with managed status and privilege level.Quote: "Identity from Entra ID and Okta: Every agent's assumed identities pulled in, with managed or unmanaged status and privilege level attached."Source: [zenity.io](https://zenity.io/platform/agentic-identity-and-access-management), read 2026-11-16. - **zenity-35** (vendor): Zenity says it uses identity to gate agent actions, such as blocking a deactivated Okta user or restricting privileged tools by role.Quote: "Zenity uses identity to gate what an agent can do, like blocking a deactivated Okta user or restricting privileged tools by role."Source: [zenity.io](https://zenity.io/platform/agentic-identity-and-access-management), read 2026-11-16. - **zenity-36** (vendor): Zenity says it flags risky non-human identities such as static API keys that coding agents use and that never appear in the IdP.Quote: "Surface risky NHIs, including static API keys and secrets that coding agents use and that never appear in your IdP."Source: [zenity.io](https://zenity.io/platform/agentic-identity-and-access-management), read 2026-11-16. - **zenity-37** (vendor): Zenity's public docs navigation lists integrations for OpenAI, Google, AWS, Salesforce, Anthropic, Cursor, GitHub, LiteLLM, Cribl Stream, Okta, CrowdStrike and Palo Alto Networks.Quote: "Gateways and custom agents LiteLLM Custom agents Cribl Stream ACS schema Identity and endpoint security Okta CrowdStrike Palo Alto Networks"Source: [docs.zenity.io](https://docs.zenity.io), read 2026-11-16. - **zenity-38** (vendor): Zenity's docs navigation lists pages for Boundaries, Inline Prevention, Kill switch, Malicious Package Protection and AI Detection & Response.Quote: "Protect Incidents AI Detection & Response Malicious Package Protection Enforce Boundaries Boundaries demos Boundaries CLI Inline Prevention Kill switch"Source: [docs.zenity.io](https://docs.zenity.io), read 2026-11-16. - **zenity-39** (none-found): None found on 2026-11-16: documentation body text readable without sign-in; docs.zenity.io/user-guide/* and /release-notes redirected to a login page, and only the navigation tree of the docs home page loaded.Source: [docs.zenity.io](https://docs.zenity.io/user-guide/inline-prevention), read 2026-11-16. - **zenity-40** (none-found): None found on 2026-11-16: documented control of browser agents, AI browsers or AI browser extensions on the platform, coding agents, MCP and AIDR pages (only a mention of agentic browsers in a press release).Source: [zenity.io](https://zenity.io/platform), read 2026-11-16. - **zenity-41** (vendor): Zenity Labs published research on zero-click attacks across agentic browsers (PleaseFix), but this is research, not a documented browser control.Quote: "Zenity Labs Exposes the Full Scope of PleaseFix, a Vulnerability Class Enabling Zero-Click Attacks Across Leading Agentic Browsers"Source: [zenity.io](https://zenity.io/company/newsroom/press-releases), read 2026-11-16. - **zenity-42** (vendor): Zenity says inventory appears the moment a platform is connected, with no separate discovery step.Quote: "Full inventory the moment a platform connects, no separate discovery step required"Source: [zenity.io](https://zenity.io/use-cases/risk-type/shadow-ai), read 2026-11-16. - **zenity-43** (vendor): Zenity says it builds a live inventory of every agent and tracks the data they access.Quote: "It builds a live inventory of every agent across an organization's environment and tracks the data those agents access"Source: [zenity.io](https://zenity.io/platform/ai-observability), read 2026-11-16. - **zenity-44** (vendor): Zenity says its posture layer evaluates agent configuration, permissions and integrations before go-live and keeps enforcing guardrails.Quote: "It evaluates configuration, permissions, and integrations before an agent goes live, and keeps enforcing guardrails automatically as agents evolve."Source: [zenity.io](https://zenity.io/platform/ai-security-posture-management), read 2026-11-16. - **zenity-45** (vendor): Zenity says its Boundaries engine evaluates every agent action in real time and can let it through, block it, or shut the agent down.Quote: "evaluates every agent action in real time against rules of an organization and decides to let it through, block it, or shut the agent down."Source: [zenity.io](https://zenity.io/platform/runtime-boundaries), read 2026-11-16. - **zenity-46** (vendor): Zenity says Boundaries stops multi-step exfiltration and privilege-escalation chains before they complete.Quote: "Multi-step exfiltration and privilege-escalation chains get stopped before they complete, not flagged after the fact."Source: [zenity.io](https://zenity.io/platform/runtime-boundaries), read 2026-11-16. - **zenity-47** (vendor): Zenity says one rule is enforced across Copilot, ChatGPT Enterprise, Agentforce, Bedrock, coding agents and homegrown builds.Quote: "Write a rule once and it's already enforced on Copilot, ChatGPT Enterprise, Agentforce, Bedrock, coding agents, and homegrown builds"Source: [zenity.io](https://zenity.io/platform/runtime-boundaries), read 2026-11-16. - **zenity-48** (vendor): Zenity gives an example of one rule blocking tool calls to specific third-party domains across every agent.Quote: "A global asset manager blocks tool calls to specific third-party domains across every agent with one rule"Source: [zenity.io](https://zenity.io/platform/runtime-boundaries), read 2026-11-16. - **zenity-49** (vendor): Zenity says a rule can be previewed to show what it would have caught before it can block anything.Quote: "See exactly what a rule would have caught before it can block anything"Source: [zenity.io](https://zenity.io/platform/runtime-boundaries), read 2026-11-16. - **zenity-50** (vendor): Zenity says that through native agent hooks and the Zenity MCP gateway it can block or modify a dangerous tool call before it executes.Quote: "it can block or modify a dangerous tool call before it executes."Source: [zenity.io](https://zenity.io/use-cases/agent-type/coding-personal-agents), read 2026-11-16. - **zenity-52** (vendor): Zenity says AIDR detects direct and indirect prompt injection and blocks execution before the agent acts on it.Quote: "AIDR detects direct and indirect prompt injection and blocks execution before the agent acts on them."Source: [zenity.io](https://zenity.io/platform/ai-detection-and-response), read 2026-11-16. - **zenity-54** (vendor): Zenity says AIDR tracks each agent step, including triggers, retrievals, tool calls and agent handoffs, in one activity table.Quote: "Step-by-step tracking across triggers, retrievals, tool invocations, and agent handoffs."Source: [zenity.io](https://zenity.io/platform/ai-detection-and-response), read 2026-11-16. - **zenity-58** (none-found): None found on 2026-11-16: a published time to deploy for the platform, searched on the platform page, product pages and press releases.Source: [zenity.io](https://zenity.io/platform), read 2026-11-16. - **zenity-59** (none-found): None found on 2026-11-16: a published price or pricing page; https://zenity.io/pricing returned HTTP 404 and the sitemap lists no pricing page.Source: [zenity.io](https://zenity.io/pricing), read 2026-11-16. - **zenity-60** (none-found): None found on 2026-11-16: documented SIEM or ticketing export on the public product pages (only Cribl Stream, Okta, CrowdStrike and Palo Alto Networks appear as integration names in the docs navigation).Source: [zenity.io](https://zenity.io/platform), read 2026-11-16. ## News mentioning Zenity - [AgentCorruption: Zenity Labs reports AWS AgentCore flaws](/news/zenity-labs-agentcorruption-aws-agentcore/) - [SalesBleed: three Agentforce flaws reported by Zenity Labs](/news/zenity-labs-salesbleed-agentforce-flaws/) ## Related pages - [All vendor profiles](/vendors/) - [WitnessAI profile](/vendors/witnessai/) - [Aim Security profile](/vendors/aim-security/) - [The ranking](/ranking/) - [How scores are set](/method/) - [Best tool by situation](/best-for/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # AI agent security solutions by layer URL: https://aiagentsecuritysolutions.com/best-for/ Updated: 2026-11-16 Five layer views re-rank the sixteen solutions for the place you secure first: endpoint, browser, cloud workspace, identity or gateway. Each view gives its own layer 30 points and keeps the other criteria as on the site ranking, so a vendor weak elsewhere can still lead one layer. ## 5 situations - [Ranked for the endpoint layer](/best-for/layer-endpoint/): Bay leads at 71.0; Bay is 1st. - [Ranked for the browser layer](/best-for/layer-browser/): Straiker leads at 65.5; Bay is 2nd. - [Ranked for the cloud workspace layer](/best-for/layer-cloud/): Operant AI and Straiker leads at 68.5; Bay is joint 5th. - [Ranked for the identity layer](/best-for/layer-identity/): CrowdStrike leads at 67.5; Bay is 4th. - [Ranked for the gateway layer](/best-for/layer-gateway/): Palo Alto Networks leads at 68.0; Bay is 16th. ## Who should pick what With no particular situation, the published ranking applies and Operant AI comes first. For a specific situation the weights change, and so can the order: - **[Ranked for the endpoint layer](/best-for/layer-endpoint/)**: pick Bay (71.0). Bay is 1st at 71.0. A buyer securing agents at the endpoint layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the browser layer](/best-for/layer-browser/)**: pick Straiker (65.5). Bay is 2nd at 65.0. A buyer securing agents at the browser layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the cloud workspace layer](/best-for/layer-cloud/)**: pick Operant AI and Straiker (68.5, joint 1st). Bay is joint 5th at 62.0. A buyer securing agents at the cloud workspace layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the identity layer](/best-for/layer-identity/)**: pick CrowdStrike (67.5). Bay is 4th at 65.0. A buyer securing agents at the identity layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. - **[Ranked for the gateway layer](/best-for/layer-gateway/)**: pick Palo Alto Networks (68.0). Bay is 16th at 47.0. A buyer securing agents at the gateway layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. ## Related pages - [The overall ranking](/ranking/) - [Weights per situation](/method/) - [Vendor profiles](/vendors/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # AI agent security solutions for the endpoint layer URL: https://aiagentsecuritysolutions.com/best-for/layer-endpoint/ Updated: 2026-11-16 Bay ranks first for the endpoint layer, which covers desktop agents, coding agents, command-line agents and local MCP servers on laptops and developer machines. The view puts 30 points on endpoint coverage and keeps the other criteria as on the site ranking. Scores were read on 16 Nov 2026. ## The short answer Bay leads the view "Ranked for the endpoint layer" at 71.0 of 100. Microsoft follows at 69.0. Bay is ahead of Microsoft mainly on stops the action (9 against 6, worth 6.0 points at weight 20). Bay is a young company that left stealth in 2026, publishes no price and has little independent testing behind its claims. ## Why these weights A buyer securing agents at the endpoint layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. **Weights used in this view** | Coverage: endpoint layer | Coverage: browser layer | Coverage: cloud workspace layer | Coverage: identity layer | Coverage: gateway layer | Stops the action | Visibility | Deployment | Policy and governance | Transparency | Proof | |---|---|---|---|---|---|---|---|---|---|---| | 30 | 0 | 0 | 0 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | **Ranked for the endpoint layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 30) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Bay](/vendors/bay/) | 71.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | 2nd | [Microsoft](/vendors/microsoft/) | 69.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 3rd | [Operant AI](/vendors/operant-ai/) | 68.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 4th | [Straiker](/vendors/straiker/) | 65.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | joint 5th | [CrowdStrike](/vendors/crowdstrike/) | 64.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | joint 5th | [Pillar Security](/vendors/pillar-security/) | 64.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 7th | [Cisco](/vendors/cisco/) | 63.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | joint 8th | [Aim Security](/vendors/aim-security/) | 62.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | joint 8th | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | joint 10th | [Lasso Security](/vendors/lasso-security/) | 62.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | joint 10th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 12th | [Noma Security](/vendors/noma-security/) | 57.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 13th | [Check Point](/vendors/check-point/) | 55.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 14th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 53.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 15th | [WitnessAI](/vendors/witnessai/) | 52.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 51.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | ## What would change this ranking **Top pair.** Microsoft rising from 8 to 9 on coverage: endpoint layer (weight 30, so each point moves a total by 3.0) would put Microsoft above Bay; the gap is 2.0 points. **Bay's next place.** Bay is already first here, so the top pair above is also the smallest change that would move Bay down. A score changes only when a sourced fact changes, and the weights were fixed before the scores. The [method page](/method/) lists the weights, anchors and rules. ## Which solutions are best for the endpoint layer? Bay ranks first for the endpoint layer on 16 Nov 2026. The view weighs endpoint coverage at 30 points and sets the other four layers to 0. Endpoint products work in the device: hooks on named coding agents, a sensor, or a client rolled out by MDM. They can block a shell command or a credential read before it runs, which no network-only product sees. ## What should the endpoint layer cover? Look for named agents, not a generic claim: Claude Code, Cursor, Codex, Gemini CLI, desktop assistants and local MCP servers. Check which operating systems are supported and whether decisions run on the device or round-trip to a cloud service. Check also how the product rolls out. Several vendors publish an MDM or EDR script as the install path, which keeps the rollout inside tools you already run. ## What is the catch with the top result? Bay is first in this view and the view's weakness line applies: Bay is a young company that left stealth in 2026, publishes no price and has little independent testing behind its claims. Bay also scores 0 on the gateway layer, which does not count here but matters if your agents reach tools through a proxy. Read Bay's profile and the next two vendors side by side before deciding. ## Related pages - [All situations](/best-for/) - [Ranked for the gateway layer](/best-for/layer-gateway/) - [Ranked for the browser layer](/best-for/layer-browser/) - [Overall ranking](/ranking/) - [Method and weights](/method/) - [Vendor profiles](/vendors/) --- # AI agent security solutions for the browser layer URL: https://aiagentsecuritysolutions.com/best-for/layer-browser/ Updated: 2026-11-16 Straiker ranks first for the browser layer, which covers browser agents, AI browsers and AI extensions. The view puts 30 points on browser coverage and keeps the other criteria as on the site ranking. Several large vendors score low here because browser agent control was none found on 16 Nov 2026. ## The short answer Straiker leads the view "Ranked for the browser layer" at 65.5 of 100. Bay is 2nd at 65.0. Straiker is ahead mainly on coverage: browser layer (7 against 6, worth 3.0 points at weight 30). ## Why these weights A buyer securing agents at the browser layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. **Weights used in this view** | Coverage: endpoint layer | Coverage: browser layer | Coverage: cloud workspace layer | Coverage: identity layer | Coverage: gateway layer | Stops the action | Visibility | Deployment | Policy and governance | Transparency | Proof | |---|---|---|---|---|---|---|---|---|---|---| | 0 | 30 | 0 | 0 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | **Ranked for the browser layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 30) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Straiker](/vendors/straiker/) | 65.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 2nd | [Bay](/vendors/bay/) | 65.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | joint 3rd | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | joint 3rd | [Operant AI](/vendors/operant-ai/) | 62.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 5th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 62.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 6th | [Aim Security](/vendors/aim-security/) | 59.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 7th | [CrowdStrike](/vendors/crowdstrike/) | 58.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 8th | [Check Point](/vendors/check-point/) | 58.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 9th | [Microsoft](/vendors/microsoft/) | 51.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 10th | [Lasso Security](/vendors/lasso-security/) | 50.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | 11th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 48.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | | 12th | [Cisco](/vendors/cisco/) | 45.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 13th | [Zenity](/vendors/zenity/) | 44.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | joint 14th | [Pillar Security](/vendors/pillar-security/) | 43.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | joint 14th | [WitnessAI](/vendors/witnessai/) | 43.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [Noma Security](/vendors/noma-security/) | 42.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | ## What would change this ranking **Top pair.** Bay rising from 6 to 7 on coverage: browser layer (weight 30, so each point moves a total by 3.0) would put Bay above Straiker; the gap is 0.5 points. **Bay's next place.** Bay rising from 6 to 7 on coverage: browser layer (weight 30, so each point moves a total by 3.0) would put Bay above Straiker; the gap is 0.5 points. A score changes only when a sourced fact changes, and the weights were fixed before the scores. The [method page](/method/) lists the weights, anchors and rules. ## Which are best for the browser layer? Straiker ranks first for the browser layer on 16 Nov 2026. The view weighs browser coverage at 30 points and sets the other four layers to 0. Browser agents act through forms, uploads, downloads, clipboard and page content, so controls must sit in the browser or an extension. Network inspection alone sees traffic but not what the agent does on the page. ## What should the browser layer cover? Separate three things: AI websites a person types into, AI extensions installed in the browser, and agents that click and fill forms on a person's behalf. Many products cover the first two, and fewer document the third. Ask which browsers are supported, whether enforcement happens in the browser or only after traffic leaves it, and whether the product can approve or deny a step instead of only logging it. ## Why do some large vendors score low here? The scores follow documentation. Where a vendor documents no control for browser agents, the browser score is 0 to 2 and the entry says none found on 16 Nov 2026. A low browser score does not mean the vendor is weak at other layers. If your agents never run in a browser, use another layer view. ## Related pages - [All situations](/best-for/) - [Ranked for the endpoint layer](/best-for/layer-endpoint/) - [Ranked for the cloud workspace layer](/best-for/layer-cloud/) - [Overall ranking](/ranking/) - [Method and weights](/method/) - [Vendor profiles](/vendors/) --- # AI agent security solutions for cloud workspaces URL: https://aiagentsecuritysolutions.com/best-for/layer-cloud/ Updated: 2026-11-16 First place for the cloud workspace layer goes to Operant AI and Straiker. The layer covers agents built or hosted in SaaS and cloud platforms such as Microsoft Copilot Studio, Salesforce Agentforce, ServiceNow and AWS Bedrock. The view puts 30 points on that layer and keeps the other criteria as on the site ranking. ## The short answer Operant AI and Straiker are joint 1st in the view "Ranked for the cloud workspace layer" at 68.5 of 100. Bay is joint 5th at 62.0. Operant AI is ahead mainly on coverage: cloud workspace layer (8 against 5, worth 9.0 points at weight 30). ## Why these weights A buyer securing agents at the cloud workspace layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. **Weights used in this view** | Coverage: endpoint layer | Coverage: browser layer | Coverage: cloud workspace layer | Coverage: identity layer | Coverage: gateway layer | Stops the action | Visibility | Deployment | Policy and governance | Transparency | Proof | |---|---|---|---|---|---|---|---|---|---|---| | 0 | 0 | 30 | 0 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | **Ranked for the cloud workspace layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 30) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | joint 1st | [Operant AI](/vendors/operant-ai/) | 68.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | joint 1st | [Straiker](/vendors/straiker/) | 68.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 3rd | [Palo Alto Networks](/vendors/palo-alto-networks/) | 68.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 4th | [Microsoft](/vendors/microsoft/) | 66.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | joint 5th | [Bay](/vendors/bay/) | 62.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | joint 5th | [Lasso Security](/vendors/lasso-security/) | 62.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | joint 5th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 8th | [Check Point](/vendors/check-point/) | 61.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 9th | [Noma Security](/vendors/noma-security/) | 60.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 10th | [Cisco](/vendors/cisco/) | 57.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 11th | [Aim Security](/vendors/aim-security/) | 53.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 12th | [Island](/vendors/island/) | 50.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | 13th | [Pillar Security](/vendors/pillar-security/) | 49.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 14th | [CrowdStrike](/vendors/crowdstrike/) | 46.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 15th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 45.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | | 16th | [WitnessAI](/vendors/witnessai/) | 43.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | ## What would change this ranking **Top pair.** Operant AI and Straiker are tied at 68.5. Straiker rising from 8 to 9 on coverage: cloud workspace layer (weight 30, so each point moves a total by 3.0) would put Straiker above Operant AI; the gap is 0.0 points. **Bay's next place.** Bay rising from 5 to 7 on coverage: cloud workspace layer (weight 30, so each point moves a total by 3.0) would put Bay above Microsoft; the gap is 4.0 points. A score changes only when a sourced fact changes, and the weights were fixed before the scores. The [method page](/method/) lists the weights, anchors and rules. ## Which are best for the cloud workspace layer? First place for the cloud workspace layer on 16 Nov 2026 goes to Operant AI and Straiker, shared where the totals match. The view weighs cloud workspace coverage at 30 points and sets the other four layers to 0. Agents built by makers inside SaaS platforms rarely run on a device you manage. Control comes from API connection to the platform, from the platform's own security features, or from a runtime hook the platform exposes. ## What should the cloud workspace layer cover? Check the platforms by name: Copilot Studio, Agentforce, ServiceNow, Bedrock and AgentCore, Vertex and Foundry. A list of integrations is a start; ask whether the product only discovers agents or can also block an action inside the platform. Some vendors discover agents through read-only API connections and enforce elsewhere. That is useful for inventory and still leaves the action itself uncontrolled. ## Does your own cloud platform already offer controls? Yes, and they come first. Microsoft documents Agent 365 with Entra identities for Copilot Studio agents, and AWS documents guardrails you can attach to a Bedrock agent. Third-party products add cross-platform inventory and one policy across several platforms. Compare what the platform gives you with what a vendor adds before you pay for overlap. Sources: [Security and governance in Microsoft Copilot Studio](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance); [Associate a guardrail with your Amazon Bedrock agent](https://docs.aws.amazon.com/bedrock/latest/userguide/agents-guardrail.html). ## Related pages - [All situations](/best-for/) - [Ranked for the browser layer](/best-for/layer-browser/) - [Ranked for the identity layer](/best-for/layer-identity/) - [Overall ranking](/ranking/) - [Method and weights](/method/) - [Vendor profiles](/vendors/) --- # AI agent security solutions for the identity layer URL: https://aiagentsecuritysolutions.com/best-for/layer-identity/ Updated: 2026-11-16 CrowdStrike ranks first for the identity layer, which covers agent identities, credentials, tokens and permissions: inventory, least privilege and revocation. The view puts 30 points on that layer and keeps the other criteria as on the site ranking. Scores were read from public documentation on 16 Nov 2026. ## The short answer CrowdStrike leads the view "Ranked for the identity layer" at 67.5 of 100. Bay is 4th at 65.0. CrowdStrike is ahead mainly on coverage: identity layer (8 against 6, worth 6.0 points at weight 30). ## Why these weights A buyer securing agents at the identity layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. **Weights used in this view** | Coverage: endpoint layer | Coverage: browser layer | Coverage: cloud workspace layer | Coverage: identity layer | Coverage: gateway layer | Stops the action | Visibility | Deployment | Policy and governance | Transparency | Proof | |---|---|---|---|---|---|---|---|---|---|---| | 0 | 0 | 0 | 30 | 0 | 20 | 15 | 10 | 10 | 10 | 5 | **Ranked for the identity layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 30) | Coverage: gateway layer (weight 0) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [CrowdStrike](/vendors/crowdstrike/) | 67.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 2nd | [Microsoft](/vendors/microsoft/) | 66.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 3rd | [Straiker](/vendors/straiker/) | 65.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 4th | [Bay](/vendors/bay/) | 65.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | | 5th | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | 6th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 7th | [Cisco](/vendors/cisco/) | 60.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 8th | [Operant AI](/vendors/operant-ai/) | 59.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 9th | [Noma Security](/vendors/noma-security/) | 57.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 10th | [Palo Alto Networks](/vendors/palo-alto-networks/) | 56.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 11th | [Lasso Security](/vendors/lasso-security/) | 53.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | 12th | [Pillar Security](/vendors/pillar-security/) | 49.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 13th | [Aim Security](/vendors/aim-security/) | 47.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 14th | [Check Point](/vendors/check-point/) | 46.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 15th | [WitnessAI](/vendors/witnessai/) | 43.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 16th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 42.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | ## What would change this ranking **Top pair.** Microsoft rising from 7 to 8 on coverage: identity layer (weight 30, so each point moves a total by 3.0) would put Microsoft above CrowdStrike; the gap is 1.5 points. **Bay's next place.** Bay rising from 6 to 7 on coverage: identity layer (weight 30, so each point moves a total by 3.0) would put Bay above Straiker; the gap is 0.5 points. A score changes only when a sourced fact changes, and the weights were fixed before the scores. The [method page](/method/) lists the weights, anchors and rules. ## Which are best for the identity layer? CrowdStrike ranks first for the identity layer on 16 Nov 2026. The view weighs identity coverage at 30 points and sets the other four layers to 0. Agent identity means knowing which agents exist, what they can reach and how to cut access. It covers inventory, scoped and short-lived credentials, and revocation. ## What should the identity layer cover? Ask for four things: an inventory of agent identities, least-privilege scoping with OAuth scopes or role limits, short-lived credentials, and a revocation path that works in minutes. Check whether the product ties into your identity provider, such as Entra or Okta. Some products only flag over-privileged access. Flagging helps with review but is not the same as issuing or revoking a credential. ## Are identity products the same as agent security solutions? No. This view scores the identity layer inside agent security solutions, not stand-alone identity or governance platforms, which are outside this site. A solution can rank here because it reads identities from your provider and gates actions on them. Pair any pick with the access controls your identity provider already offers, such as conditional access policy. Sources: [Microsoft Entra Conditional Access overview](https://learn.microsoft.com/en-us/entra/identity/conditional-access/overview). ## Related pages - [All situations](/best-for/) - [Ranked for the cloud workspace layer](/best-for/layer-cloud/) - [Ranked for the gateway layer](/best-for/layer-gateway/) - [Overall ranking](/ranking/) - [Method and weights](/method/) - [Vendor profiles](/vendors/) --- # AI agent security solutions for the gateway layer URL: https://aiagentsecuritysolutions.com/best-for/layer-gateway/ Updated: 2026-11-16 Palo Alto Networks ranks first for the gateway layer, a proxy or API control point between agents and models, tools or MCP servers. The view puts 30 points on that layer. Bay says it decides on the device with no proxy; none found on 16 Nov 2026 for a gateway, so it sits last here. ## The short answer Palo Alto Networks leads the view "Ranked for the gateway layer" at 68.0 of 100. Bay is 16th at 47.0. Palo Alto Networks is ahead mainly on coverage: gateway layer (8 against 0, worth 24.0 points at weight 30). ## Why these weights A buyer securing agents at the gateway layer first weighs coverage of that layer as all five layer weights together, and keeps the other criteria as on the site ranking. **Weights used in this view** | Coverage: endpoint layer | Coverage: browser layer | Coverage: cloud workspace layer | Coverage: identity layer | Coverage: gateway layer | Stops the action | Visibility | Deployment | Policy and governance | Transparency | Proof | |---|---|---|---|---|---|---|---|---|---|---| | 0 | 0 | 0 | 0 | 30 | 20 | 15 | 10 | 10 | 10 | 5 | **Ranked for the gateway layer: scores out of 10 per criterion, totals out of 100** | Rank | Vendor | Total (of 100) | Coverage: endpoint layer (weight 0) | Coverage: browser layer (weight 0) | Coverage: cloud workspace layer (weight 0) | Coverage: identity layer (weight 0) | Coverage: gateway layer (weight 30) | Stops the action (weight 20) | Visibility (weight 15) | Deployment (weight 10) | Policy and governance (weight 10) | Transparency (weight 10) | Proof (weight 5) | |---|---|---|---|---|---|---|---|---|---|---|---|---|---| | 1st | [Palo Alto Networks](/vendors/palo-alto-networks/) | 68.0 | [3](/vendors/palo-alto-networks/#evidence-endpoint) | [6](/vendors/palo-alto-networks/#evidence-browser) | [8](/vendors/palo-alto-networks/#evidence-cloud) | [4](/vendors/palo-alto-networks/#evidence-identity) | [8](/vendors/palo-alto-networks/#evidence-gateway) | [6](/vendors/palo-alto-networks/#evidence-stops) | [6](/vendors/palo-alto-networks/#evidence-visibility) | [7](/vendors/palo-alto-networks/#evidence-deploy) | [5](/vendors/palo-alto-networks/#evidence-policy) | [7](/vendors/palo-alto-networks/#evidence-transparency) | [8](/vendors/palo-alto-networks/#evidence-proof) | | 2nd | [Operant AI](/vendors/operant-ai/) | 65.5 | [8](/vendors/operant-ai/#evidence-endpoint) | [6](/vendors/operant-ai/#evidence-browser) | [8](/vendors/operant-ai/#evidence-cloud) | [5](/vendors/operant-ai/#evidence-identity) | [7](/vendors/operant-ai/#evidence-gateway) | [7](/vendors/operant-ai/#evidence-stops) | [7](/vendors/operant-ai/#evidence-visibility) | [7](/vendors/operant-ai/#evidence-deploy) | [7](/vendors/operant-ai/#evidence-policy) | [3](/vendors/operant-ai/#evidence-transparency) | [6](/vendors/operant-ai/#evidence-proof) | | 3rd | [Lasso Security](/vendors/lasso-security/) | 65.0 | [6](/vendors/lasso-security/#evidence-endpoint) | [2](/vendors/lasso-security/#evidence-browser) | [6](/vendors/lasso-security/#evidence-cloud) | [3](/vendors/lasso-security/#evidence-identity) | [7](/vendors/lasso-security/#evidence-gateway) | [7](/vendors/lasso-security/#evidence-stops) | [7](/vendors/lasso-security/#evidence-visibility) | [7](/vendors/lasso-security/#evidence-deploy) | [6](/vendors/lasso-security/#evidence-policy) | [3](/vendors/lasso-security/#evidence-transparency) | [7](/vendors/lasso-security/#evidence-proof) | | 4th | [CrowdStrike](/vendors/crowdstrike/) | 64.5 | [7](/vendors/crowdstrike/#evidence-endpoint) | [5](/vendors/crowdstrike/#evidence-browser) | [1](/vendors/crowdstrike/#evidence-cloud) | [8](/vendors/crowdstrike/#evidence-identity) | [7](/vendors/crowdstrike/#evidence-gateway) | [7](/vendors/crowdstrike/#evidence-stops) | [7](/vendors/crowdstrike/#evidence-visibility) | [6](/vendors/crowdstrike/#evidence-deploy) | [6](/vendors/crowdstrike/#evidence-policy) | [3](/vendors/crowdstrike/#evidence-transparency) | [8](/vendors/crowdstrike/#evidence-proof) | | 5th | [Microsoft](/vendors/microsoft/) | 63.0 | [8](/vendors/microsoft/#evidence-endpoint) | [2](/vendors/microsoft/#evidence-browser) | [7](/vendors/microsoft/#evidence-cloud) | [7](/vendors/microsoft/#evidence-identity) | [6](/vendors/microsoft/#evidence-gateway) | [6](/vendors/microsoft/#evidence-stops) | [7](/vendors/microsoft/#evidence-visibility) | [7](/vendors/microsoft/#evidence-deploy) | [6](/vendors/microsoft/#evidence-policy) | [7](/vendors/microsoft/#evidence-transparency) | [5](/vendors/microsoft/#evidence-proof) | | 6th | [Island](/vendors/island/) | 62.5 | [7](/vendors/island/#evidence-endpoint) | [7](/vendors/island/#evidence-browser) | [3](/vendors/island/#evidence-cloud) | [7](/vendors/island/#evidence-identity) | [7](/vendors/island/#evidence-gateway) | [7](/vendors/island/#evidence-stops) | [7](/vendors/island/#evidence-visibility) | [5](/vendors/island/#evidence-deploy) | [5](/vendors/island/#evidence-policy) | [3](/vendors/island/#evidence-transparency) | [8](/vendors/island/#evidence-proof) | | 7th | [Zenity](/vendors/zenity/) | 62.0 | [7](/vendors/zenity/#evidence-endpoint) | [1](/vendors/zenity/#evidence-browser) | [7](/vendors/zenity/#evidence-cloud) | [7](/vendors/zenity/#evidence-identity) | [7](/vendors/zenity/#evidence-gateway) | [7](/vendors/zenity/#evidence-stops) | [7](/vendors/zenity/#evidence-visibility) | [5](/vendors/zenity/#evidence-deploy) | [5](/vendors/zenity/#evidence-policy) | [3](/vendors/zenity/#evidence-transparency) | [7](/vendors/zenity/#evidence-proof) | | 8th | [WitnessAI](/vendors/witnessai/) | 61.5 | [5](/vendors/witnessai/#evidence-endpoint) | [2](/vendors/witnessai/#evidence-browser) | [2](/vendors/witnessai/#evidence-cloud) | [2](/vendors/witnessai/#evidence-identity) | [8](/vendors/witnessai/#evidence-gateway) | [7](/vendors/witnessai/#evidence-stops) | [6](/vendors/witnessai/#evidence-visibility) | [5](/vendors/witnessai/#evidence-deploy) | [4](/vendors/witnessai/#evidence-policy) | [2](/vendors/witnessai/#evidence-transparency) | [7](/vendors/witnessai/#evidence-proof) | | 9th | [Check Point](/vendors/check-point/) | 61.0 | [4](/vendors/check-point/#evidence-endpoint) | [5](/vendors/check-point/#evidence-browser) | [6](/vendors/check-point/#evidence-cloud) | [1](/vendors/check-point/#evidence-identity) | [6](/vendors/check-point/#evidence-gateway) | [6](/vendors/check-point/#evidence-stops) | [7](/vendors/check-point/#evidence-visibility) | [6](/vendors/check-point/#evidence-deploy) | [4](/vendors/check-point/#evidence-policy) | [7](/vendors/check-point/#evidence-transparency) | [7](/vendors/check-point/#evidence-proof) | | 10th | [Cisco](/vendors/cisco/) | 60.0 | [7](/vendors/cisco/#evidence-endpoint) | [1](/vendors/cisco/#evidence-browser) | [5](/vendors/cisco/#evidence-cloud) | [6](/vendors/cisco/#evidence-identity) | [6](/vendors/cisco/#evidence-gateway) | [7](/vendors/cisco/#evidence-stops) | [5](/vendors/cisco/#evidence-visibility) | [6](/vendors/cisco/#evidence-deploy) | [6](/vendors/cisco/#evidence-policy) | [6](/vendors/cisco/#evidence-transparency) | [5](/vendors/cisco/#evidence-proof) | | 11th | [Aim Security](/vendors/aim-security/) | 59.5 | [7](/vendors/aim-security/#evidence-endpoint) | [6](/vendors/aim-security/#evidence-browser) | [4](/vendors/aim-security/#evidence-cloud) | [2](/vendors/aim-security/#evidence-identity) | [6](/vendors/aim-security/#evidence-gateway) | [7](/vendors/aim-security/#evidence-stops) | [7](/vendors/aim-security/#evidence-visibility) | [5](/vendors/aim-security/#evidence-deploy) | [5](/vendors/aim-security/#evidence-policy) | [3](/vendors/aim-security/#evidence-transparency) | [8](/vendors/aim-security/#evidence-proof) | | 12th | [Pillar Security](/vendors/pillar-security/) | 58.5 | [7](/vendors/pillar-security/#evidence-endpoint) | [0](/vendors/pillar-security/#evidence-browser) | [2](/vendors/pillar-security/#evidence-cloud) | [2](/vendors/pillar-security/#evidence-identity) | [5](/vendors/pillar-security/#evidence-gateway) | [7](/vendors/pillar-security/#evidence-stops) | [7](/vendors/pillar-security/#evidence-visibility) | [6](/vendors/pillar-security/#evidence-deploy) | [7](/vendors/pillar-security/#evidence-policy) | [3](/vendors/pillar-security/#evidence-transparency) | [6](/vendors/pillar-security/#evidence-proof) | | 13th | [Noma Security](/vendors/noma-security/) | 57.0 | [6](/vendors/noma-security/#evidence-endpoint) | [1](/vendors/noma-security/#evidence-browser) | [7](/vendors/noma-security/#evidence-cloud) | [6](/vendors/noma-security/#evidence-identity) | [6](/vendors/noma-security/#evidence-gateway) | [6](/vendors/noma-security/#evidence-stops) | [7](/vendors/noma-security/#evidence-visibility) | [5](/vendors/noma-security/#evidence-deploy) | [5](/vendors/noma-security/#evidence-policy) | [3](/vendors/noma-security/#evidence-transparency) | [7](/vendors/noma-security/#evidence-proof) | | 14th | [Straiker](/vendors/straiker/) | 53.5 | [7](/vendors/straiker/#evidence-endpoint) | [7](/vendors/straiker/#evidence-browser) | [8](/vendors/straiker/#evidence-cloud) | [7](/vendors/straiker/#evidence-identity) | [3](/vendors/straiker/#evidence-gateway) | [7](/vendors/straiker/#evidence-stops) | [7](/vendors/straiker/#evidence-visibility) | [8](/vendors/straiker/#evidence-deploy) | [5](/vendors/straiker/#evidence-policy) | [3](/vendors/straiker/#evidence-transparency) | [8](/vendors/straiker/#evidence-proof) | | 15th | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | 48.0 | [5](/vendors/sentinelone/#evidence-endpoint) | [4](/vendors/sentinelone/#evidence-browser) | [3](/vendors/sentinelone/#evidence-cloud) | [2](/vendors/sentinelone/#evidence-identity) | [4](/vendors/sentinelone/#evidence-gateway) | [5](/vendors/sentinelone/#evidence-stops) | [7](/vendors/sentinelone/#evidence-visibility) | [5](/vendors/sentinelone/#evidence-deploy) | [4](/vendors/sentinelone/#evidence-policy) | [3](/vendors/sentinelone/#evidence-transparency) | [7](/vendors/sentinelone/#evidence-proof) | | 16th | [Bay](/vendors/bay/) | 47.0 | [8](/vendors/bay/#evidence-endpoint) | [6](/vendors/bay/#evidence-browser) | [5](/vendors/bay/#evidence-cloud) | [6](/vendors/bay/#evidence-identity) | [0](/vendors/bay/#evidence-gateway) | [9](/vendors/bay/#evidence-stops) | [7](/vendors/bay/#evidence-visibility) | [8](/vendors/bay/#evidence-deploy) | [6](/vendors/bay/#evidence-policy) | [1](/vendors/bay/#evidence-transparency) | [7](/vendors/bay/#evidence-proof) | ## What would change this ranking **Top pair.** Operant AI rising from 7 to 8 on coverage: gateway layer (weight 30, so each point moves a total by 3.0) would put Operant AI above Palo Alto Networks; the gap is 2.5 points. **Bay's next place.** Bay rising from 0 to 1 on coverage: gateway layer (weight 30, so each point moves a total by 3.0) would put Bay above SentinelOne (Prompt Security); the gap is 1.0 points. A score changes only when a sourced fact changes, and the weights were fixed before the scores. The [method page](/method/) lists the weights, anchors and rules. ## Which are best for the gateway layer? Palo Alto Networks ranks first for the gateway layer on 16 Nov 2026. The view weighs gateway coverage at 30 points and sets the other four layers to 0. A gateway sits on the traffic path, so it sees every call that is routed through it and can block, mask or log before the call completes. It sees nothing that bypasses it. ## What should the gateway layer cover? Check whether the vendor runs its own gateway, plugs into one you already run (Kong, Envoy, LiteLLM and similar), or enforces on the network. Check whether MCP calls and tool calls pass through it as well as model prompts. Ask what happens when the gateway fails. Cisco documents a fail-closed default for its Kong plugin; other vendors publish nothing on the point. ## What if a solution has no gateway? Then it acts somewhere else, and that can be a better fit. Bay says it decides on the device with no proxy, and with none found on 16 Nov 2026 for a gateway it scores 0 on this layer, and a team with managed laptops may prefer that. A team whose agents run on servers and call tools over the network needs a control point on that path. Use the Agent Control Point Map to see the moment each vendor acts. ## Related pages - [All situations](/best-for/) - [Ranked for the identity layer](/best-for/layer-identity/) - [Ranked for the endpoint layer](/best-for/layer-endpoint/) - [Overall ranking](/ranking/) - [Method and weights](/method/) - [Vendor profiles](/vendors/) --- # Layer Coverage Map for AI agent security URL: https://aiagentsecuritysolutions.com/layer-coverage-map/ Updated: 2026-11-16 The Layer Coverage Map records which of 16 vendors document AI agent security at five layers: endpoint, browser, cloud workspace, identity and gateway. It holds 49 yes cells out of 80, each tied to dated facts read on 16 Nov 2026. 12 vendors have a yes on three or more layers. ## The matrix **Layer Coverage Map: each vendor against each dimension** | Vendor | [Endpoint](/layer-coverage-map/endpoint/) | [Browser](/layer-coverage-map/browser/) | [Cloud workspace](/layer-coverage-map/cloud-workspace/) | [Identity](/layer-coverage-map/identity/) | [Gateway](/layer-coverage-map/gateway/) | |---|---|---|---|---|---| | [Aim Security](/vendors/aim-security/) | [Yes](/vendors/aim-security/#fact-aim-security-11) | [Yes](/vendors/aim-security/#fact-aim-security-42) | [Partial](/vendors/aim-security/#fact-aim-security-44) | [Partial](/vendors/aim-security/#fact-aim-security-47) | [Yes](/vendors/aim-security/#fact-aim-security-49) | | [Bay](/vendors/bay/) | [Yes](/vendors/bay/#fact-bay-14) | [Yes](/vendors/bay/#fact-bay-12) | [Partial](/vendors/bay/#fact-bay-37) | [Partial](/vendors/bay/#fact-bay-41) | [None found](/vendors/bay/#fact-bay-44) | | [Check Point](/vendors/check-point/) | [Partial](/vendors/check-point/#fact-check-point-08) | [Yes](/vendors/check-point/#fact-check-point-12) | [Yes](/vendors/check-point/#fact-check-point-15) | [None found](/vendors/check-point/#fact-check-point-19) | [Yes](/vendors/check-point/#fact-check-point-22) | | [Cisco](/vendors/cisco/) | [Yes](/vendors/cisco/#fact-cisco-10) | [None found](/vendors/cisco/#fact-cisco-47) | [Partial](/vendors/cisco/#fact-cisco-43) | [Partial](/vendors/cisco/#fact-cisco-39) | [Yes](/vendors/cisco/#fact-cisco-05) | | [CrowdStrike](/vendors/crowdstrike/) | [Yes](/vendors/crowdstrike/#fact-crowdstrike-02) | [Partial](/vendors/crowdstrike/#fact-crowdstrike-35) | [Partial](/vendors/crowdstrike/#fact-crowdstrike-26) | [Yes](/vendors/crowdstrike/#fact-crowdstrike-31) | [Yes](/vendors/crowdstrike/#fact-crowdstrike-18) | | [Island](/vendors/island/) | [Yes](/vendors/island/#fact-island-04) | [Yes](/vendors/island/#fact-island-07) | [Partial](/vendors/island/#fact-island-10) | [Yes](/vendors/island/#fact-island-12) | [Yes](/vendors/island/#fact-island-15) | | [Lasso Security](/vendors/lasso-security/) | [Yes](/vendors/lasso-security/#fact-lasso-security-19) | [Partial](/vendors/lasso-security/#fact-lasso-security-36) | [Yes](/vendors/lasso-security/#fact-lasso-security-32) | [Partial](/vendors/lasso-security/#fact-lasso-security-34) | [Yes](/vendors/lasso-security/#fact-lasso-security-16) | | [Microsoft](/vendors/microsoft/) | [Yes](/vendors/microsoft/#fact-microsoft-11) | [Partial](/vendors/microsoft/#fact-microsoft-08) | [Yes](/vendors/microsoft/#fact-microsoft-31) | [Yes](/vendors/microsoft/#fact-microsoft-33) | [Partial](/vendors/microsoft/#fact-microsoft-28) | | [Noma Security](/vendors/noma-security/) | [Yes](/vendors/noma-security/#fact-noma-security-08) | [None found](/vendors/noma-security/#fact-noma-security-10) | [Yes](/vendors/noma-security/#fact-noma-security-12) | [Yes](/vendors/noma-security/#fact-noma-security-15) | [Yes](/vendors/noma-security/#fact-noma-security-18) | | [Operant AI](/vendors/operant-ai/) | [Yes](/vendors/operant-ai/#fact-operant-ai-09) | [Yes](/vendors/operant-ai/#fact-operant-ai-12) | [Yes](/vendors/operant-ai/#fact-operant-ai-14) | [Yes](/vendors/operant-ai/#fact-operant-ai-18) | [Yes](/vendors/operant-ai/#fact-operant-ai-21) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | [Partial](/vendors/palo-alto-networks/#fact-palo-alto-networks-04) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-08) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-33) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-38) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-41) | | [Pillar Security](/vendors/pillar-security/) | [Yes](/vendors/pillar-security/#fact-pillar-security-05) | [None found](/vendors/pillar-security/#fact-pillar-security-45) | [None found](/vendors/pillar-security/#fact-pillar-security-46) | [Partial](/vendors/pillar-security/#fact-pillar-security-16) | [Yes](/vendors/pillar-security/#fact-pillar-security-27) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | [Yes](/vendors/sentinelone/#fact-sentinelone-13) | [Yes](/vendors/sentinelone/#fact-sentinelone-13) | [Partial](/vendors/sentinelone/#fact-sentinelone-34) | [Partial](/vendors/sentinelone/#fact-sentinelone-38) | [Yes](/vendors/sentinelone/#fact-sentinelone-14) | | [Straiker](/vendors/straiker/) | [Yes](/vendors/straiker/#fact-straiker-06) | [Yes](/vendors/straiker/#fact-straiker-09) | [Yes](/vendors/straiker/#fact-straiker-12) | [Partial](/vendors/straiker/#fact-straiker-15) | [Partial](/vendors/straiker/#fact-straiker-19) | | [WitnessAI](/vendors/witnessai/) | [Partial](/vendors/witnessai/#fact-witnessai-25) | [Partial](/vendors/witnessai/#fact-witnessai-28) | [Partial](/vendors/witnessai/#fact-witnessai-29) | [Partial](/vendors/witnessai/#fact-witnessai-31) | [Yes](/vendors/witnessai/#fact-witnessai-33) | | [Zenity](/vendors/zenity/) | [Yes](/vendors/zenity/#fact-zenity-19) | [None found](/vendors/zenity/#fact-zenity-40) | [Yes](/vendors/zenity/#fact-zenity-30) | [Yes](/vendors/zenity/#fact-zenity-34) | [Yes](/vendors/zenity/#fact-zenity-24) | - **Yes**: documented specifically - **Partial**: documented in general terms, for some agents only, or detect and alert without blocking - **None found**: none found on 16 Nov 2026 ## What the map shows Agents run in more than one place, and a control at one layer is blind to the others. The map lists 16 vendors against five layers and gives each cell one of three levels, with the fact ids behind it and a one-sentence note. Read it by layer to see who documents the place your agents run, and by vendor to see how many layers one product covers. A yes records documentation, and the map does not test performance. ## Which products cover several layers? 12 of 16 vendors have a yes on three or more of the five layers on 16 Nov 2026: Aim Security, Check Point, CrowdStrike, Island, Lasso Security, Microsoft, Noma Security, Operant AI, Palo Alto Networks, SentinelOne (Prompt Security), Straiker, Zenity. Yes on all five layers: Operant AI. Yes on four layers: Island, Noma Security, Palo Alto Networks, Zenity. Yes on exactly three layers: Aim Security, Check Point, CrowdStrike, Lasso Security, Microsoft, SentinelOne (Prompt Security), Straiker. The other 4 vendors have a yes on two layers or fewer. These counts are computed from the cells and count only yes cells. A partial cell is not counted, because partial means documented in general terms, for some agents only, or detect and alert without blocking. A product that covers many layers still needs checking layer by layer, since the depth behind each yes differs. ## How to read a cell Each cell has a level, a note and the fact ids that support it. The note is the evidence pass's own one-sentence summary. Where a layer shows none found, the pass looked and found nothing on 16 Nov 2026, which is not proof that a capability is absent. ## Download and cite [Download the Layer Coverage Map as CSV or JSON](/datasets/layer-coverage-map/). Cite it as: AI Agent Security Solutions, edition 16 Nov 2026, https://aiagentsecuritysolutions.com/, CC BY 4.0. ## Related pages - [How the Layer Coverage Map is built](/layer-coverage-map/methodology/) - [Layer Coverage Map data download](/datasets/layer-coverage-map/) - [Endpoint](/layer-coverage-map/endpoint/) - [Browser](/layer-coverage-map/browser/) - [Cloud workspace](/layer-coverage-map/cloud-workspace/) - [Identity](/layer-coverage-map/identity/) - [Gateway](/layer-coverage-map/gateway/) - [Agent Control Point Map](/agent-control-point-map/) - [Vendor profiles](/vendors/) - [The ranking](/ranking/) --- # How the Layer Coverage Map was built URL: https://aiagentsecuritysolutions.com/layer-coverage-map/methodology/ Updated: 2026-11-16 The Layer Coverage Map records, for 16 vendors, what each documents across 5 dimensions. Levels were set in one evidence pass on 16 Nov 2026 from vendor pages and third-party reports, with three levels: yes, partial and none found. Every cell cites fact ids, and each fact carries a URL and read date. ## What the Layer Coverage Map is The Layer Coverage Map is a table of 16 vendors against five layers where AI agents run: the endpoint, the browser, cloud workspaces and SaaS platforms, the identity system, and gateways between agents and models or tools. Each cell shows what the vendor documents at that layer. ## How levels were set in the 16 Nov 2026 evidence pass The evidence pass read each vendor's product pages, documentation, release notes, published research and at least two third-party sources, and recorded dated facts with a URL, a quote and a read date. Each cell then took one of three levels from those facts, using the same definitions for every vendor. Yes means documented specifically. Partial means documented in general terms, for some agents or clients only, or detect and alert without blocking. None found means nothing was found on 16 Nov 2026, which records the state of the evidence and is not a claim about the product. Vendor statements are written as the vendor's. Nothing was run in a lab for this edition, so a yes records what a vendor documents and not how well it performs. - yes: documented specifically - partial: documented in general terms, for some agents only, or detect and alert without blocking - none-found: none found on 16 Nov 2026 ## Sources Each cell lists the fact ids that support it. The facts, with URLs, quotes and read dates, are in the vendor evidence files shown on each vendor profile and in the downloadable JSON on the datasets page. Pages that blocked a plain fetch were read through a text reader and cited by their original address. ## How to cite Cite as: AI Agent Security Solutions, Layer Coverage Map, edition of 16 Nov 2026, https://aiagentsecuritysolutions.com/layer-coverage-map/, CC BY 4.0. Cite a single cell with the vendor, the dimension and the edition date. ## License The Layer Coverage Map and the Agent Control Point Map are published under the Creative Commons Attribution 4.0 license (CC BY 4.0): copy, share and adapt them with credit. ## Corrections If a cell is wrong or a vendor has published newer documentation, send the page URL through the corrections page. Verified changes are dated and logged on that page, and the edition date is updated when cells change. ## Levels and counts - **Yes** (49 cells): documented specifically - **Partial** (24 cells): documented in general terms, for some agents only, or detect and alert without blocking - **None found** (7 cells): none found on 16 Nov 2026 ## Related pages - [The Layer Coverage Map](/layer-coverage-map/) - [Layer Coverage Map download](/datasets/layer-coverage-map/) - [Scoring method](/method/) - [Endpoint](/layer-coverage-map/endpoint/) - [Browser](/layer-coverage-map/browser/) - [Cloud workspace](/layer-coverage-map/cloud-workspace/) --- # AI agent security at the endpoint layer URL: https://aiagentsecuritysolutions.com/layer-coverage-map/endpoint/ Updated: 2026-11-16 Endpoint coverage means a vendor documents controls for desktop agents, coding agents, command-line agents and local MCP servers on employee machines. On 16 Nov 2026, 13 of 16 vendors had a yes at the endpoint layer, 3 had partial and 0 had none found. Levels are from vendor documentation, not lab tests. ## Endpoint: every vendor **Endpoint by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | Yes | Local agents are found and tool calls blocked on the endpoint. | [aim-security-11](/vendors/aim-security/#fact-aim-security-11), [aim-security-40](/vendors/aim-security/#fact-aim-security-40), [aim-security-41](/vendors/aim-security/#fact-aim-security-41) | | [Bay](/vendors/bay/) | Yes | Endpoint product covers coding agents, desktop assistants and MCP servers. | [bay-14](/vendors/bay/#fact-bay-14), [bay-72](/vendors/bay/#fact-bay-72), [bay-73](/vendors/bay/#fact-bay-73) | | [Cisco](/vendors/cisco/) | Yes | DefenseClaw hooks for named coding agents on Linux, macOS and Windows. | [cisco-10](/vendors/cisco/#fact-cisco-10), [cisco-11](/vendors/cisco/#fact-cisco-11), [cisco-12](/vendors/cisco/#fact-cisco-12) | | [CrowdStrike](/vendors/crowdstrike/) | Yes | Falcon sensor visibility plus agent-level hooks for Claude Code, Codex, Copilot, OpenCode and Cursor on Windows, macOS and Linux. | [crowdstrike-02](/vendors/crowdstrike/#fact-crowdstrike-02), [crowdstrike-09](/vendors/crowdstrike/#fact-crowdstrike-09), [crowdstrike-10](/vendors/crowdstrike/#fact-crowdstrike-10) | | [Island](/vendors/island/) | Yes | Island Desktop inventories and governs agents, MCP servers and skills on managed devices. | [island-04](/vendors/island/#fact-island-04), [island-05](/vendors/island/#fact-island-05) | | [Lasso Security](/vendors/lasso-security/) | Yes | Lasso says it inventories Claude Code, Cursor and Windsurf connections and governs endpoint AI agentlessly. | [lasso-security-19](/vendors/lasso-security/#fact-lasso-security-19), [lasso-security-26](/vendors/lasso-security/#fact-lasso-security-26), [lasso-security-12](/vendors/lasso-security/#fact-lasso-security-12) | | [Microsoft](/vendors/microsoft/) | Yes | Defender for Endpoint discovers and protects local agents. | [microsoft-11](/vendors/microsoft/#fact-microsoft-11), [microsoft-19](/vendors/microsoft/#fact-microsoft-19) | | [Noma Security](/vendors/noma-security/) | Yes | Endpoint discovery via EDR or MDM and named coding agents. | [noma-security-08](/vendors/noma-security/#fact-noma-security-08), [noma-security-09](/vendors/noma-security/#fact-noma-security-09) | | [Operant AI](/vendors/operant-ai/) | Yes | Endpoint Protector inspects coding agents, MCP clients and plugins on macOS, Windows, Linux. | [operant-ai-09](/vendors/operant-ai/#fact-operant-ai-09), [operant-ai-10](/vendors/operant-ai/#fact-operant-ai-10) | | [Pillar Security](/vendors/pillar-security/) | Yes | Coding agent discovery, posture and pre-action blocking on managed machines. | [pillar-security-05](/vendors/pillar-security/#fact-pillar-security-05), [pillar-security-09](/vendors/pillar-security/#fact-pillar-security-09) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Yes | Dedicated AI endpoint agent that detects new agents and enforces runtime policy. | [sentinelone-13](/vendors/sentinelone/#fact-sentinelone-13), [sentinelone-32](/vendors/sentinelone/#fact-sentinelone-32) | | [Straiker](/vendors/straiker/) | Yes | Coding agents such as Cursor, Copilot, Claude Code and Windsurf covered via a thin client; endpoint discovery stated. | [straiker-06](/vendors/straiker/#fact-straiker-06), [straiker-07](/vendors/straiker/#fact-straiker-07) | | [Zenity](/vendors/zenity/) | Yes | Coding and personal agents on laptops and desktops are discovered and blocked via native hooks and MCP gateway. | [zenity-19](/vendors/zenity/#fact-zenity-19), [zenity-50](/vendors/zenity/#fact-zenity-50) | | [Check Point](/vendors/check-point/) | Partial | Desktop agent and IDE coverage is claimed under Workforce AI Security; OS list and local control detail none found. | [check-point-08](/vendors/check-point/#fact-check-point-08), [check-point-10](/vendors/check-point/#fact-check-point-10), [check-point-11](/vendors/check-point/#fact-check-point-11) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Partial | Only community example hooks for coding assistants in a best-effort GitHub repo; endpoint governance is in a separate profile. | [palo-alto-networks-04](/vendors/palo-alto-networks/#fact-palo-alto-networks-04), [palo-alto-networks-50](/vendors/palo-alto-networks/#fact-palo-alto-networks-50) | | [WitnessAI](/vendors/witnessai/) | Partial | Sees developer-machine agents through network traffic and Cursor, Claude Code and Codex workflows; no endpoint agent documented. | [witnessai-25](/vendors/witnessai/#fact-witnessai-25), [witnessai-26](/vendors/witnessai/#fact-witnessai-26) | ## What agents run at the endpoint layer Coding agents, desktop assistants, command-line agents and locally installed MCP servers run as processes on employee laptops. Cisco says an interactive coding agent can read a repository, run the build and push a change, and acts with the full access rights of the developer who started it. Noma says such agents, MCP servers and skills carry the same permissions and credentials as the people who installed them. Sources: [Protect enterprise coding agents with Cisco DefenseClaw and NVIDIA OpenShell](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell). ## What a control at the endpoint layer can and cannot see A control on the machine can see local processes, commands, configuration files, skills directories and file access, and it can act before a local action completes. Noma says endpoint activity leaves local evidence in configuration files, skills directories, connector history and running processes. It cannot see agents that run in a vendor's cloud. Operant notes that very little Copilot activity has a local process to instrument, because the prompt is answered with content pulled from cloud services the user never opened on the machine. Sources: [Noma Extends Agent Security and Governance to the Employee Endpoint](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html); [Operant: coverage for Microsoft Copilot and embedded AI assistants](https://www.operant.ai/art-kubed/announcing-operants-coverage-for-microsoft-copilot-and-every-ai-assistant-embedded-in-microsoft-365). ## Who documents the endpoint layer Yes on 16 Nov 2026: Aim Security, Bay, Cisco, CrowdStrike, Island, Lasso Security, Microsoft, Noma Security, Operant AI, Pillar Security, SentinelOne (Prompt Security), Straiker, Zenity. Partial: Check Point, Palo Alto Networks, WitnessAI. None found: none. ## Related pages - [The full Layer Coverage Map](/layer-coverage-map/) - [Gateway](/layer-coverage-map/gateway/) - [Browser](/layer-coverage-map/browser/) - [Method](/layer-coverage-map/methodology/) - [Vendor profiles](/vendors/) --- # AI agent security at the browser layer URL: https://aiagentsecuritysolutions.com/layer-coverage-map/browser/ Updated: 2026-11-16 Browser coverage means a vendor documents controls for AI browsers, AI extensions and agents that act inside web pages. On 16 Nov 2026, 8 of 16 vendors had a yes at the browser layer, 4 had partial and 4 had none found. Levels are from vendor documentation, not lab tests. ## Browser: every vendor **Browser by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | Yes | Browser AI prompts can be monitored, anonymized or blocked. | [aim-security-42](/vendors/aim-security/#fact-aim-security-42), [aim-security-43](/vendors/aim-security/#fact-aim-security-43) | | [Bay](/vendors/bay/) | Yes | Browser product covers AI tools, extensions and browser agents. | [bay-12](/vendors/bay/#fact-bay-12), [bay-24](/vendors/bay/#fact-bay-24), [bay-74](/vendors/bay/#fact-bay-74) | | [Check Point](/vendors/check-point/) | Yes | Browser extension is the base deployment and covers web AI apps and browser extensions. | [check-point-12](/vendors/check-point/#fact-check-point-12), [check-point-14](/vendors/check-point/#fact-check-point-14) | | [Island](/vendors/island/) | Yes | Governed headless browser for agents plus extension support for AI browsers. | [island-07](/vendors/island/#fact-island-07), [island-08](/vendors/island/#fact-island-08) | | [Operant AI](/vendors/operant-ai/) | Yes | Browser Protection for AI browsers and gen AI sites, announced 2026-08-27. | [operant-ai-12](/vendors/operant-ai/#fact-operant-ai-12), [operant-ai-13](/vendors/operant-ai/#fact-operant-ai-13) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | Prisma Browser governs GenAI apps, extensions and agentic browsing. | [palo-alto-networks-08](/vendors/palo-alto-networks/#fact-palo-alto-networks-08), [palo-alto-networks-09](/vendors/palo-alto-networks/#fact-palo-alto-networks-09), [palo-alto-networks-49](/vendors/palo-alto-networks/#fact-palo-alto-networks-49) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Yes | Browser extension with prompt inspection and redaction. | [sentinelone-13](/vendors/sentinelone/#fact-sentinelone-13), [sentinelone-11](/vendors/sentinelone/#fact-sentinelone-11) | | [Straiker](/vendors/straiker/) | Yes | Agentic browser guardrails and a Browser Guard plus extension are documented. | [straiker-09](/vendors/straiker/#fact-straiker-09), [straiker-10](/vendors/straiker/#fact-straiker-10) | | [CrowdStrike](/vendors/crowdstrike/) | Partial | Seraphic Enterprise Browser (separate product) inventories extensions and controls paste and upload to generative AI tools; no browser-agent control documented. | [crowdstrike-35](/vendors/crowdstrike/#fact-crowdstrike-35), [crowdstrike-36](/vendors/crowdstrike/#fact-crowdstrike-36), [crowdstrike-37](/vendors/crowdstrike/#fact-crowdstrike-37) | | [Lasso Security](/vendors/lasso-security/) | Partial | A browser extension is named as a deployment path, but no browser-agent product page found. | [lasso-security-36](/vendors/lasso-security/#fact-lasso-security-36), [lasso-security-37](/vendors/lasso-security/#fact-lasso-security-37) | | [Microsoft](/vendors/microsoft/) | Partial | Purview browser extension and Endpoint DLP paste control, no AI browser agent control found. | [microsoft-08](/vendors/microsoft/#fact-microsoft-08), [microsoft-10](/vendors/microsoft/#fact-microsoft-10), [microsoft-40](/vendors/microsoft/#fact-microsoft-40), [microsoft-67](/vendors/microsoft/#fact-microsoft-67) | | [WitnessAI](/vendors/witnessai/) | Partial | Covers VS Code AI extensions and ChatGPT plugins; no AI-browser coverage found. | [witnessai-28](/vendors/witnessai/#fact-witnessai-28), [witnessai-27](/vendors/witnessai/#fact-witnessai-27) | | [Cisco](/vendors/cisco/) | None found | No browser-agent coverage found; AI Access inspects web traffic. | [cisco-47](/vendors/cisco/#fact-cisco-47), [cisco-48](/vendors/cisco/#fact-cisco-48) | | [Noma Security](/vendors/noma-security/) | None found | No browser-layer coverage documented. | [noma-security-10](/vendors/noma-security/#fact-noma-security-10), [noma-security-11](/vendors/noma-security/#fact-noma-security-11) | | [Pillar Security](/vendors/pillar-security/) | None found | None found on 2026-11-16. | [pillar-security-45](/vendors/pillar-security/#fact-pillar-security-45) | | [Zenity](/vendors/zenity/) | None found | No browser-agent control documented; only research on agentic browsers and a mention in a press release. | [zenity-40](/vendors/zenity/#fact-zenity-40), [zenity-41](/vendors/zenity/#fact-zenity-41) | ## What agents run at the browser layer AI browsers, AI extensions and browser agents read pages and act on a user's behalf inside the browser. Island says its extension supports AI browsers including Atlas and Comet with the same feature support as in Chrome and Edge. CrowdStrike says its Seraphic product works in existing browsers without a dedicated browser or traffic rerouting. Sources: [Island: enterprise browser for agents](https://www.island.io/product/ai/enterprise-browser-for-agents). ## What a control at the browser layer can and cannot see A control in the browser can see page content, pastes, uploads, downloads and the extension list, and can allow or block extensions and data movement. CrowdStrike says it can inventory extensions, rate their risk and enforce copy, paste, upload and download policies for generative AI tools. It cannot see desktop applications or command-line agents. Cato says browser extensions alone cannot secure AI interactions in desktop applications or through agents running locally on an endpoint. Sources: [CrowdStrike Brings Enterprise Browser Security to the Falcon Platform](https://www.crowdstrike.com/en-us/blog/crowdstrike-brings-enterprise-browser-security-to-falcon-platform/); [Cato Networks Launches Cato AI Endpoint](https://www.catonetworks.com/news/cato-launches-cato-ai-endpoint-secure-ai-use-beyond-browser/). ## Who documents the browser layer Yes on 16 Nov 2026: Aim Security, Bay, Check Point, Island, Operant AI, Palo Alto Networks, SentinelOne (Prompt Security), Straiker. Partial: CrowdStrike, Lasso Security, Microsoft, WitnessAI. None found: Cisco, Noma Security, Pillar Security, Zenity. ## Related pages - [The full Layer Coverage Map](/layer-coverage-map/) - [Endpoint](/layer-coverage-map/endpoint/) - [Cloud workspace](/layer-coverage-map/cloud-workspace/) - [Method](/layer-coverage-map/methodology/) - [Vendor profiles](/vendors/) --- # AI agent security at the cloud workspace layer URL: https://aiagentsecuritysolutions.com/layer-coverage-map/cloud-workspace/ Updated: 2026-11-16 Cloud workspace coverage means a vendor documents controls for agents built or hosted in SaaS and cloud platforms such as Copilot Studio, Agentforce or AWS agent services. On 16 Nov 2026, 8 of 16 vendors had a yes at the cloud workspace layer, 7 had partial and 1 had none found. Levels are from vendor documentation, not lab tests. ## Cloud workspace: every vendor **Cloud workspace by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Check Point](/vendors/check-point/) | Yes | Discovery connectors named for Bedrock, AgentCore, Google Cloud, Copilot Studio, Agentforce, n8n and Relevance AI. | [check-point-15](/vendors/check-point/#fact-check-point-15), [check-point-16](/vendors/check-point/#fact-check-point-16) | | [Lasso Security](/vendors/lasso-security/) | Yes | Documented discovery across Vertex AI, Copilot, Bedrock and Agentforce plus Claude Compliance API. | [lasso-security-32](/vendors/lasso-security/#fact-lasso-security-32), [lasso-security-33](/vendors/lasso-security/#fact-lasso-security-33), [lasso-security-45](/vendors/lasso-security/#fact-lasso-security-45) | | [Microsoft](/vendors/microsoft/) | Yes | Copilot Studio, Foundry, Bedrock and Vertex agents in inventory. | [microsoft-31](/vendors/microsoft/#fact-microsoft-31), [microsoft-32](/vendors/microsoft/#fact-microsoft-32) | | [Noma Security](/vendors/noma-security/) | Yes | SaaS and cloud agent platforms named and covered by API. | [noma-security-12](/vendors/noma-security/#fact-noma-security-12), [noma-security-13](/vendors/noma-security/#fact-noma-security-13), [noma-security-14](/vendors/noma-security/#fact-noma-security-14) | | [Operant AI](/vendors/operant-ai/) | Yes | Agent Protector covers cloud and SaaS agents, Microsoft 365 Agents SDK, AWS and Azure. | [operant-ai-14](/vendors/operant-ai/#fact-operant-ai-14), [operant-ai-15](/vendors/operant-ai/#fact-operant-ai-15), [operant-ai-16](/vendors/operant-ai/#fact-operant-ai-16) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | SaaS Agent Security covers Copilot Studio and ServiceNow agents; agent discovery covers onboarded cloud accounts. | [palo-alto-networks-33](/vendors/palo-alto-networks/#fact-palo-alto-networks-33), [palo-alto-networks-34](/vendors/palo-alto-networks/#fact-palo-alto-networks-34), [palo-alto-networks-35](/vendors/palo-alto-networks/#fact-palo-alto-networks-35) | | [Straiker](/vendors/straiker/) | Yes | Copilot Studio, Agentforce, Bedrock AgentCore and Azure AI Foundry named. | [straiker-12](/vendors/straiker/#fact-straiker-12), [straiker-13](/vendors/straiker/#fact-straiker-13) | | [Zenity](/vendors/zenity/) | Yes | Covers Copilot Studio, Agentforce, ChatGPT Enterprise, Bedrock AgentCore, Foundry and Vertex AI. | [zenity-30](/vendors/zenity/#fact-zenity-30), [zenity-31](/vendors/zenity/#fact-zenity-31), [zenity-32](/vendors/zenity/#fact-zenity-32), [zenity-47](/vendors/zenity/#fact-zenity-47) | | [Aim Security](/vendors/aim-security/) | Partial | Amazon Bedrock workloads via API and managed agents in general terms; named SaaS agent platforms not found. | [aim-security-44](/vendors/aim-security/#fact-aim-security-44), [aim-security-45](/vendors/aim-security/#fact-aim-security-45), [aim-security-46](/vendors/aim-security/#fact-aim-security-46) | | [Bay](/vendors/bay/) | Partial | VMs, Kubernetes and Coder workspaces documented; named SaaS agent platforms none found. | [bay-37](/vendors/bay/#fact-bay-37), [bay-38](/vendors/bay/#fact-bay-38), [bay-39](/vendors/bay/#fact-bay-39), [bay-40](/vendors/bay/#fact-bay-40) | | [Cisco](/vendors/cisco/) | Partial | Cloud AI inventory and AgentCore/ADK inspection; no SaaS agent platforms named. | [cisco-43](/vendors/cisco/#fact-cisco-43), [cisco-44](/vendors/cisco/#fact-cisco-44), [cisco-45](/vendors/cisco/#fact-cisco-45), [cisco-46](/vendors/cisco/#fact-cisco-46) | | [CrowdStrike](/vendors/crowdstrike/) | Partial | Cloud workload, Kubernetes and Bedrock AgentCore coverage is pre-beta or coming soon. | [crowdstrike-26](/vendors/crowdstrike/#fact-crowdstrike-26), [crowdstrike-27](/vendors/crowdstrike/#fact-crowdstrike-27), [crowdstrike-28](/vendors/crowdstrike/#fact-crowdstrike-28) | | [Island](/vendors/island/) | Partial | Claude surfaces read through the Compliance API; no Copilot Studio, Agentforce, ServiceNow or Bedrock agent coverage found. | [island-10](/vendors/island/#fact-island-10), [island-05](/vendors/island/#fact-island-05) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Partial | Amazon Bedrock AgentCore only; Copilot Studio, Agentforce, ServiceNow not found. | [sentinelone-34](/vendors/sentinelone/#fact-sentinelone-34), [sentinelone-36](/vendors/sentinelone/#fact-sentinelone-36), [sentinelone-37](/vendors/sentinelone/#fact-sentinelone-37) | | [WitnessAI](/vendors/witnessai/) | Partial | Discovers custom agents built in the public cloud; no named SaaS agent platforms found. | [witnessai-29](/vendors/witnessai/#fact-witnessai-29), [witnessai-30](/vendors/witnessai/#fact-witnessai-30) | | [Pillar Security](/vendors/pillar-security/) | None found | No named SaaS or cloud agent platform coverage found. | [pillar-security-46](/vendors/pillar-security/#fact-pillar-security-46) | ## What agents run at the cloud workspace layer Agents built or hosted in SaaS and cloud platforms run in the platform vendor's cloud, with that platform's data and identity model. Zenity Labs reported flaws in Salesforce Agentforce in September 2026 and in Amazon Bedrock AgentCore in October 2026, both agent platforms of this kind. Operant lists Copilot in Word and Teams, Copilot Studio agents and Agents SDK agents as separate entry points. Sources: [SalesBleed: 0-click data exfiltration on Agentforce](https://labs.zenity.io/post/salesbleed-0-click-data-exfiltration-on-agentforce). ## What a control at the cloud workspace layer can and cannot see A control here sees records, tool calls, conversations, memories and outbound destinations through the platform's logs or an API integration, and it can enforce inline only where the platform exposes a hook. Operant says its Purview path detects and alerts but does not sit inline, while its Agent Protector evaluates each step of agents built on the Agents SDK. No process runs on the employee machine, so endpoint and browser controls see nothing of what these agents do. Zenity says one over-broad default role in AgentCore reached every agent in an account and region, which shows why platform configuration is part of this layer. Sources: [Zenity Labs reports AgentCorruption](https://zenity.io/press-release/zenity-labs-discloses-agentcorruption-a-chain-of-aws-agentcore-flaws); [Operant: coverage for Microsoft Copilot and embedded AI assistants](https://www.operant.ai/art-kubed/announcing-operants-coverage-for-microsoft-copilot-and-every-ai-assistant-embedded-in-microsoft-365). ## Who documents the cloud workspace layer Yes on 16 Nov 2026: Check Point, Lasso Security, Microsoft, Noma Security, Operant AI, Palo Alto Networks, Straiker, Zenity. Partial: Aim Security, Bay, Cisco, CrowdStrike, Island, SentinelOne (Prompt Security), WitnessAI. None found: Pillar Security. ## Related pages - [The full Layer Coverage Map](/layer-coverage-map/) - [Browser](/layer-coverage-map/browser/) - [Identity](/layer-coverage-map/identity/) - [Method](/layer-coverage-map/methodology/) - [Vendor profiles](/vendors/) --- # AI agent security at the identity layer URL: https://aiagentsecuritysolutions.com/layer-coverage-map/identity/ Updated: 2026-11-16 Identity coverage means a vendor documents controls for agent identities, credentials, tokens and permissions, including inventory, least privilege and revocation. On 16 Nov 2026, 7 of 16 vendors had a yes at the identity layer, 8 had partial and 1 had none found. Levels are from vendor documentation, not lab tests. ## Identity: every vendor **Identity by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [CrowdStrike](/vendors/crowdstrike/) | Yes | Agentic Identity Provider registers agents, issues verifiable identities and brokers short-lived scoped access. | [crowdstrike-31](/vendors/crowdstrike/#fact-crowdstrike-31), [crowdstrike-32](/vendors/crowdstrike/#fact-crowdstrike-32), [crowdstrike-33](/vendors/crowdstrike/#fact-crowdstrike-33) | | [Island](/vendors/island/) | Yes | Non-human identity inventory, granted versus used access, scoped credentials. | [island-12](/vendors/island/#fact-island-12), [island-14](/vendors/island/#fact-island-14) | | [Microsoft](/vendors/microsoft/) | Yes | Entra Agent ID and Conditional Access for agents. | [microsoft-33](/vendors/microsoft/#fact-microsoft-33), [microsoft-34](/vendors/microsoft/#fact-microsoft-34) | | [Noma Security](/vendors/noma-security/) | Yes | Agent identity attribution and owner mapping documented. | [noma-security-15](/vendors/noma-security/#fact-noma-security-15), [noma-security-16](/vendors/noma-security/#fact-noma-security-16), [noma-security-17](/vendors/noma-security/#fact-noma-security-17) | | [Operant AI](/vendors/operant-ai/) | Yes | Agent identity catalog and access enforcement; MCP non-human identities. | [operant-ai-18](/vendors/operant-ai/#fact-operant-ai-18), [operant-ai-19](/vendors/operant-ai/#fact-operant-ai-19), [operant-ai-20](/vendors/operant-ai/#fact-operant-ai-20) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | Agent Identity Security and Idira agent identities are documented. | [palo-alto-networks-38](/vendors/palo-alto-networks/#fact-palo-alto-networks-38), [palo-alto-networks-39](/vendors/palo-alto-networks/#fact-palo-alto-networks-39), [palo-alto-networks-40](/vendors/palo-alto-networks/#fact-palo-alto-networks-40) | | [Zenity](/vendors/zenity/) | Yes | Pulls agent identities from Entra ID and Okta, flags non-human identities and gates actions by identity. | [zenity-34](/vendors/zenity/#fact-zenity-34), [zenity-35](/vendors/zenity/#fact-zenity-35), [zenity-36](/vendors/zenity/#fact-zenity-36) | | [Aim Security](/vendors/aim-security/) | Partial | Over-privileged access is flagged; agent identity and token management not documented. | [aim-security-47](/vendors/aim-security/#fact-aim-security-47), [aim-security-48](/vendors/aim-security/#fact-aim-security-48) | | [Bay](/vendors/bay/) | Partial | Credential exposure, permission visibility and attribution documented; agent identity lifecycle and revocation not found. | [bay-41](/vendors/bay/#fact-bay-41), [bay-42](/vendors/bay/#fact-bay-42), [bay-43](/vendors/bay/#fact-bay-43) | | [Cisco](/vendors/cisco/) | Partial | Duo IAM agent registration is planned; Privileged Infrastructure Access launched. | [cisco-39](/vendors/cisco/#fact-cisco-39), [cisco-42](/vendors/cisco/#fact-cisco-42), [cisco-55](/vendors/cisco/#fact-cisco-55) | | [Lasso Security](/vendors/lasso-security/) | Partial | Flags overprivileged agents and role-based access to MCP servers; no credential or revocation controls found. | [lasso-security-34](/vendors/lasso-security/#fact-lasso-security-34), [lasso-security-24](/vendors/lasso-security/#fact-lasso-security-24), [lasso-security-35](/vendors/lasso-security/#fact-lasso-security-35) | | [Pillar Security](/vendors/pillar-security/) | Partial | Posture issues cover identity and RBAC for AI assets, but no agent credential revocation found. | [pillar-security-16](/vendors/pillar-security/#fact-pillar-security-16), [pillar-security-23](/vendors/pillar-security/#fact-pillar-security-23) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Partial | Least-privilege scoping stated; credential and token inventory or revocation not found. | [sentinelone-38](/vendors/sentinelone/#fact-sentinelone-38), [sentinelone-39](/vendors/sentinelone/#fact-sentinelone-39) | | [Straiker](/vendors/straiker/) | Partial | Permissions and OAuth scopes are mapped and access can be revoked, but no agent identity inventory is documented. | [straiker-15](/vendors/straiker/#fact-straiker-15), [straiker-16](/vendors/straiker/#fact-straiker-16), [straiker-17](/vendors/straiker/#fact-straiker-17) | | [WitnessAI](/vendors/witnessai/) | Partial | Links agent activity to the human identity; no agent credential or token inventory found. | [witnessai-31](/vendors/witnessai/#fact-witnessai-31), [witnessai-32](/vendors/witnessai/#fact-witnessai-32) | | [Check Point](/vendors/check-point/) | None found | No agent identity, token or revocation feature found; docs defer identity to other products. | [check-point-19](/vendors/check-point/#fact-check-point-19), [check-point-20](/vendors/check-point/#fact-check-point-20) | ## What agents run at the identity layer Agents hold or borrow credentials: API keys, OAuth tokens, service accounts and delegated user access. Microsoft says every agent should have a first-class identity distinct from the person using or sponsoring it, and CrowdStrike says its identity provider registers agents in one directory and brokers short-lived, scoped access. OWASP's non-human identity list names long-lived secrets among its ten risks. Sources: [Agents are your newest insiders. Security must move at their speed.](https://www.microsoft.com/en-us/security/blog/2026/10/19/agents-are-your-newest-insiders-security-must-move-at-their-speed/). ## What a control at the identity layer can and cannot see An identity control can see which agents are registered, what each may access, and when a token was issued, and it can narrow or revoke access without touching the agent itself. Microsoft says Entra Agent ID gives agents the same identity-driven protections as users, including lifecycle management. It cannot see what an agent does with access once granted, or agents that never register. Microsoft says it is adding identity types so local and previously unknown agents found by sensors can become policy targets, which acknowledges that discovery has to come first. Sources: [CrowdStrike Announces Agentic Identity Provider](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/); [OWASP Non-Human Identity Top 10 2025](https://owasp.org/www-project-non-human-identities-top-10/2025/). ## Who documents the identity layer Yes on 16 Nov 2026: CrowdStrike, Island, Microsoft, Noma Security, Operant AI, Palo Alto Networks, Zenity. Partial: Aim Security, Bay, Cisco, Lasso Security, Pillar Security, SentinelOne (Prompt Security), Straiker, WitnessAI. None found: Check Point. ## Related pages - [The full Layer Coverage Map](/layer-coverage-map/) - [Cloud workspace](/layer-coverage-map/cloud-workspace/) - [Gateway](/layer-coverage-map/gateway/) - [Method](/layer-coverage-map/methodology/) - [Vendor profiles](/vendors/) --- # AI agent security at the gateway layer URL: https://aiagentsecuritysolutions.com/layer-coverage-map/gateway/ Updated: 2026-11-16 Gateway coverage means a vendor documents a proxy or gateway between agents and models, tools or MCP servers that inspects and enforces policy on the traffic. On 16 Nov 2026, 13 of 16 vendors had a yes at the gateway layer, 2 had partial and 1 had none found. Levels are from vendor documentation, not lab tests. ## Gateway: every vendor **Gateway by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | Yes | Proxy, AI gateway or API integration points are documented. | [aim-security-49](/vendors/aim-security/#fact-aim-security-49), [aim-security-50](/vendors/aim-security/#fact-aim-security-50) | | [Check Point](/vendors/check-point/) | Yes | Guard API and inline enforcement screen agent prompts, outputs and tool calls; native platform integrations are roadmap. | [check-point-22](/vendors/check-point/#fact-check-point-22), [check-point-23](/vendors/check-point/#fact-check-point-23) | | [Cisco](/vendors/cisco/) | Yes | Kong plugin inspects LLM and MCP traffic; Cisco MCP gateway described as in development. | [cisco-05](/vendors/cisco/#fact-cisco-05), [cisco-07](/vendors/cisco/#fact-cisco-07), [cisco-41](/vendors/cisco/#fact-cisco-41) | | [CrowdStrike](/vendors/crowdstrike/) | Yes | Falcon AI Gateway, included with Guardian, is generally available and drives the AI Kill Switch. | [crowdstrike-18](/vendors/crowdstrike/#fact-crowdstrike-18), [crowdstrike-23](/vendors/crowdstrike/#fact-crowdstrike-23), [crowdstrike-24](/vendors/crowdstrike/#fact-crowdstrike-24) | | [Island](/vendors/island/) | Yes | Island MCP Gateway sits between agents and tools; LLM gateway and inference hook integrations. | [island-15](/vendors/island/#fact-island-15), [island-13](/vendors/island/#fact-island-13) | | [Lasso Security](/vendors/lasso-security/) | Yes | Works with Kong, Envoy, TrueFoundry, Portkey and LiteLLM and ships an open-source MCP gateway. | [lasso-security-16](/vendors/lasso-security/#fact-lasso-security-16), [lasso-security-14](/vendors/lasso-security/#fact-lasso-security-14), [lasso-security-31](/vendors/lasso-security/#fact-lasso-security-31) | | [Noma Security](/vendors/noma-security/) | Yes | Enforces through AI and MCP gateways, hooks, SDKs and APIs. | [noma-security-18](/vendors/noma-security/#fact-noma-security-18), [noma-security-19](/vendors/noma-security/#fact-noma-security-19) | | [Operant AI](/vendors/operant-ai/) | Yes | MCP Gateway trust zones and AI Gatekeeper LLM gateway. | [operant-ai-21](/vendors/operant-ai/#fact-operant-ai-21), [operant-ai-22](/vendors/operant-ai/#fact-operant-ai-22) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | AI Gateway proxies LLM calls with inline guardrails; an MCP server tool scans tool calls. | [palo-alto-networks-41](/vendors/palo-alto-networks/#fact-palo-alto-networks-41), [palo-alto-networks-42](/vendors/palo-alto-networks/#fact-palo-alto-networks-42), [palo-alto-networks-45](/vendors/palo-alto-networks/#fact-palo-alto-networks-45) | | [Pillar Security](/vendors/pillar-security/) | Yes | Runtime guardrails work with named AI gateways and any gateway via API. | [pillar-security-27](/vendors/pillar-security/#fact-pillar-security-27) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Yes | MCP Gateway and enforcement at the AgentCore gateway. | [sentinelone-14](/vendors/sentinelone/#fact-sentinelone-14), [sentinelone-35](/vendors/sentinelone/#fact-sentinelone-35) | | [WitnessAI](/vendors/witnessai/) | Yes | Network enforcement of an approved MCP server and tool list, plus REST API protection in the orchestration layer. | [witnessai-33](/vendors/witnessai/#fact-witnessai-33), [witnessai-34](/vendors/witnessai/#fact-witnessai-34), [witnessai-36](/vendors/witnessai/#fact-witnessai-36) | | [Zenity](/vendors/zenity/) | Yes | MCP gateway URL with allow, modify or block of tool calls. | [zenity-24](/vendors/zenity/#fact-zenity-24), [zenity-25](/vendors/zenity/#fact-zenity-25) | | [Microsoft](/vendors/microsoft/) | Partial | Open-source MCP Gateway and Agent 365 MCP tool real-time protection. | [microsoft-28](/vendors/microsoft/#fact-microsoft-28), [microsoft-39](/vendors/microsoft/#fact-microsoft-39), [microsoft-68](/vendors/microsoft/#fact-microsoft-68) | | [Straiker](/vendors/straiker/) | Partial | Integrates with AI gateways; a Straiker gateway itself is not documented. | [straiker-19](/vendors/straiker/#fact-straiker-19), [straiker-20](/vendors/straiker/#fact-straiker-20), [straiker-21](/vendors/straiker/#fact-straiker-21) | | [Bay](/vendors/bay/) | None found | No gateway or proxy product found; vendor says it uses no proxy. | [bay-44](/vendors/bay/#fact-bay-44), [bay-45](/vendors/bay/#fact-bay-45) | ## What agents run at the gateway layer A gateway sits between applications or agents and the models, tools or MCP servers they call. CrowdStrike says developers connect an existing SDK or coding tool by changing the base URL to the gateway and using a gateway key. Check Point and Clalit Health Services announced on 1 October 2026 that they will jointly build a dedicated AI gateway between users, applications and AI models. Sources: [Inside Falcon AI Gateway: One Control Point for AI Traffic](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/). ## What a control at the gateway layer can and cannot see A gateway can see prompts, responses and usage on the traffic that reaches it, and can enforce one policy for managed and unmanaged devices. CrowdStrike says Falcon AI Gateway inspects prompts and responses and enforces policy across both. It cannot see traffic that bypasses it, or what an agent does on a machine after a response arrives. CrowdStrike discussed keeping the request path short but gave no latency number or fail-open or fail-closed behavior in its post, so those remain open questions for buyers. Sources: [Inside Falcon AI Gateway: One Control Point for AI Traffic](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/); [Clalit Health Services and Check Point announce a strategic partnership](https://www.checkpoint.com/press-releases/clalit-health-services-and-check-point-announce-strategic-partnership-to-secure-ai-adoption-across-healthcare-environments/). ## Who documents the gateway layer Yes on 16 Nov 2026: Aim Security, Check Point, Cisco, CrowdStrike, Island, Lasso Security, Noma Security, Operant AI, Palo Alto Networks, Pillar Security, SentinelOne (Prompt Security), WitnessAI, Zenity. Partial: Microsoft, Straiker. None found: Bay. ## Related pages - [The full Layer Coverage Map](/layer-coverage-map/) - [Identity](/layer-coverage-map/identity/) - [Endpoint](/layer-coverage-map/endpoint/) - [Method](/layer-coverage-map/methodology/) - [Vendor profiles](/vendors/) --- # Agent Control Point Map for AI agent security URL: https://aiagentsecuritysolutions.com/agent-control-point-map/ Updated: 2026-11-16 The Agent Control Point Map records which of 16 vendors document five controls for AI agents: inventory, posture assessment, runtime block, permissions and revocation, and audit trail. Each of the 80 cells carries a level, a note and fact ids from evidence read on 16 Nov 2026. ## The matrix **Agent Control Point Map: each vendor against each dimension** | Vendor | [Agent inventory](/agent-control-point-map/agent-inventory/) | [Posture assessment](/agent-control-point-map/posture-assessment/) | [Runtime block](/agent-control-point-map/runtime-block/) | [Permissions and revocation](/agent-control-point-map/permissions-and-revocation/) | [Audit trail](/agent-control-point-map/audit-trail/) | |---|---|---|---|---|---| | [Aim Security](/vendors/aim-security/) | [Yes](/vendors/aim-security/#fact-aim-security-54) | [Yes](/vendors/aim-security/#fact-aim-security-27) | [Yes](/vendors/aim-security/#fact-aim-security-51) | [Partial](/vendors/aim-security/#fact-aim-security-47) | [Yes](/vendors/aim-security/#fact-aim-security-52) | | [Bay](/vendors/bay/) | [Yes](/vendors/bay/#fact-bay-11) | [Yes](/vendors/bay/#fact-bay-66) | [Yes](/vendors/bay/#fact-bay-47) | [Partial](/vendors/bay/#fact-bay-31) | [Yes](/vendors/bay/#fact-bay-61) | | [Check Point](/vendors/check-point/) | [Yes](/vendors/check-point/#fact-check-point-15) | [Yes](/vendors/check-point/#fact-check-point-21) | [Yes](/vendors/check-point/#fact-check-point-25) | [Partial](/vendors/check-point/#fact-check-point-27) | [Partial](/vendors/check-point/#fact-check-point-36) | | [Cisco](/vendors/cisco/) | [Partial](/vendors/cisco/#fact-cisco-42) | [Partial](/vendors/cisco/#fact-cisco-02) | [Yes](/vendors/cisco/#fact-cisco-10) | [Partial](/vendors/cisco/#fact-cisco-16) | [Yes](/vendors/cisco/#fact-cisco-17) | | [CrowdStrike](/vendors/crowdstrike/) | [Yes](/vendors/crowdstrike/#fact-crowdstrike-01) | [Partial](/vendors/crowdstrike/#fact-crowdstrike-01) | [Yes](/vendors/crowdstrike/#fact-crowdstrike-10) | [Yes](/vendors/crowdstrike/#fact-crowdstrike-31) | [Yes](/vendors/crowdstrike/#fact-crowdstrike-38) | | [Island](/vendors/island/) | [Yes](/vendors/island/#fact-island-04) | [Yes](/vendors/island/#fact-island-06) | [Yes](/vendors/island/#fact-island-16) | [Yes](/vendors/island/#fact-island-13) | [Yes](/vendors/island/#fact-island-19) | | [Lasso Security](/vendors/lasso-security/) | [Yes](/vendors/lasso-security/#fact-lasso-security-32) | [Yes](/vendors/lasso-security/#fact-lasso-security-54) | [Yes](/vendors/lasso-security/#fact-lasso-security-10) | [Partial](/vendors/lasso-security/#fact-lasso-security-24) | [Yes](/vendors/lasso-security/#fact-lasso-security-25) | | [Microsoft](/vendors/microsoft/) | [Yes](/vendors/microsoft/#fact-microsoft-31) | [Yes](/vendors/microsoft/#fact-microsoft-29) | [Yes](/vendors/microsoft/#fact-microsoft-28) | [Yes](/vendors/microsoft/#fact-microsoft-34) | [Yes](/vendors/microsoft/#fact-microsoft-37) | | [Noma Security](/vendors/noma-security/) | [Yes](/vendors/noma-security/#fact-noma-security-25) | [Yes](/vendors/noma-security/#fact-noma-security-47) | [Yes](/vendors/noma-security/#fact-noma-security-20) | [Yes](/vendors/noma-security/#fact-noma-security-23) | [Yes](/vendors/noma-security/#fact-noma-security-31) | | [Operant AI](/vendors/operant-ai/) | [Yes](/vendors/operant-ai/#fact-operant-ai-14) | [Partial](/vendors/operant-ai/#fact-operant-ai-39) | [Yes](/vendors/operant-ai/#fact-operant-ai-23) | [Yes](/vendors/operant-ai/#fact-operant-ai-19) | [Yes](/vendors/operant-ai/#fact-operant-ai-37) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-35) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-33) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-34) | [Partial](/vendors/palo-alto-networks/#fact-palo-alto-networks-39) | [Yes](/vendors/palo-alto-networks/#fact-palo-alto-networks-36) | | [Pillar Security](/vendors/pillar-security/) | [Yes](/vendors/pillar-security/#fact-pillar-security-04) | [Yes](/vendors/pillar-security/#fact-pillar-security-08) | [Yes](/vendors/pillar-security/#fact-pillar-security-09) | [Partial](/vendors/pillar-security/#fact-pillar-security-14) | [Yes](/vendors/pillar-security/#fact-pillar-security-12) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | [Yes](/vendors/sentinelone/#fact-sentinelone-10) | [Partial](/vendors/sentinelone/#fact-sentinelone-45) | [Yes](/vendors/sentinelone/#fact-sentinelone-43) | [Partial](/vendors/sentinelone/#fact-sentinelone-38) | [Yes](/vendors/sentinelone/#fact-sentinelone-44) | | [Straiker](/vendors/straiker/) | [Yes](/vendors/straiker/#fact-straiker-06) | [Yes](/vendors/straiker/#fact-straiker-16) | [Yes](/vendors/straiker/#fact-straiker-22) | [Partial](/vendors/straiker/#fact-straiker-15) | [Yes](/vendors/straiker/#fact-straiker-27) | | [WitnessAI](/vendors/witnessai/) | [Yes](/vendors/witnessai/#fact-witnessai-37) | [Partial](/vendors/witnessai/#fact-witnessai-17) | [Yes](/vendors/witnessai/#fact-witnessai-14) | [Partial](/vendors/witnessai/#fact-witnessai-34) | [Yes](/vendors/witnessai/#fact-witnessai-40) | | [Zenity](/vendors/zenity/) | [Yes](/vendors/zenity/#fact-zenity-43) | [Yes](/vendors/zenity/#fact-zenity-44) | [Yes](/vendors/zenity/#fact-zenity-45) | [Partial](/vendors/zenity/#fact-zenity-35) | [Yes](/vendors/zenity/#fact-zenity-27) | - **Yes**: documented specifically - **Partial**: documented in general terms, for some agents only, or detect and alert without blocking - **None found**: none found on 16 Nov 2026 ## What the map shows The Layer Coverage Map asks where a vendor works. This map asks what the vendor does there: find agents, assess them, stop actions, limit access and keep a record. Read the columns as a sequence. Inventory finds the agent, posture checks its setup, permissions limit it, runtime block stops a bad action, and the audit trail shows what happened. A gap at one point is not covered by strength at another. ## How to use the map when comparing vendors Check the control points that match your first incident worry. If you worry about an agent being hijacked mid-task, runtime block and permissions matter most. If you worry about unknown agents, start with inventory. Where a cell reads partial, open the note: it says what was documented and what was not. A partial often means detect and alert without blocking. ## Download and cite [Download the Agent Control Point Map as CSV or JSON](/datasets/agent-control-point-map/). Cite it as: AI Agent Security Solutions, edition 16 Nov 2026, https://aiagentsecuritysolutions.com/, CC BY 4.0. ## Related pages - [How the Agent Control Point Map is built](/agent-control-point-map/methodology/) - [Agent Control Point Map data download](/datasets/agent-control-point-map/) - [Agent inventory](/agent-control-point-map/agent-inventory/) - [Posture assessment](/agent-control-point-map/posture-assessment/) - [Runtime block](/agent-control-point-map/runtime-block/) - [Permissions and revocation](/agent-control-point-map/permissions-and-revocation/) - [Audit trail](/agent-control-point-map/audit-trail/) - [Layer Coverage Map](/layer-coverage-map/) - [Vendor profiles](/vendors/) - [The ranking](/ranking/) --- # How the Agent Control Point Map was built URL: https://aiagentsecuritysolutions.com/agent-control-point-map/methodology/ Updated: 2026-11-16 The Agent Control Point Map records, for 16 vendors, what each documents across 5 dimensions. Levels were set in one evidence pass on 16 Nov 2026 from vendor pages and third-party reports, with three levels: yes, partial and none found. Every cell cites fact ids, and each fact carries a URL and read date. ## What the Agent Control Point Map is The Agent Control Point Map is a table of 16 vendors against five controls that apply to AI agents: inventory, posture assessment, runtime block, permissions and revocation, and audit trail. Each cell shows what the vendor documents for that control. ## How levels were set in the 16 Nov 2026 evidence pass The evidence pass read each vendor's product pages, documentation, release notes, published research and at least two third-party sources, and recorded dated facts with a URL, a quote and a read date. Each cell then took one of three levels from those facts, using the same definitions for every vendor. Yes means documented specifically. Partial means documented in general terms, for some agents or clients only, or detect and alert without blocking. None found means nothing was found on 16 Nov 2026, which records the state of the evidence and is not a claim about the product. Vendor statements are written as the vendor's. Nothing was run in a lab for this edition, so a yes records what a vendor documents and not how well it performs. - yes: documented specifically - partial: documented in general terms, for some agents only, or detect and alert without blocking - none-found: none found on 16 Nov 2026 ## Sources Each cell lists the fact ids that support it. The facts, with URLs, quotes and read dates, are in the vendor evidence files shown on each vendor profile and in the downloadable JSON on the datasets page. Pages that blocked a plain fetch were read through a text reader and cited by their original address. ## How to cite Cite as: AI Agent Security Solutions, Agent Control Point Map, edition of 16 Nov 2026, https://aiagentsecuritysolutions.com/agent-control-point-map/, CC BY 4.0. Cite a single cell with the vendor, the dimension and the edition date. ## License The Layer Coverage Map and the Agent Control Point Map are published under the Creative Commons Attribution 4.0 license (CC BY 4.0): copy, share and adapt them with credit. ## Corrections If a cell is wrong or a vendor has published newer documentation, send the page URL through the corrections page. Verified changes are dated and logged on that page, and the edition date is updated when cells change. ## Levels and counts - **Yes** (62 cells): documented specifically - **Partial** (18 cells): documented in general terms, for some agents only, or detect and alert without blocking - **None found** (0 cells): none found on 16 Nov 2026 ## Related pages - [The Agent Control Point Map](/agent-control-point-map/) - [Agent Control Point Map download](/datasets/agent-control-point-map/) - [Scoring method](/method/) - [Agent inventory](/agent-control-point-map/agent-inventory/) - [Posture assessment](/agent-control-point-map/posture-assessment/) - [Runtime block](/agent-control-point-map/runtime-block/) --- # Agent inventory as an AI agent control point URL: https://aiagentsecuritysolutions.com/agent-control-point-map/agent-inventory/ Updated: 2026-11-16 Agent inventory is the control that finds and lists the agents, MCP servers and skills in use. On 16 Nov 2026, 15 of 16 vendors had a yes for this control point, 1 had partial and 0 had none found. Levels are from vendor documentation read that day, not lab tests. ## Agent inventory: every vendor **Agent inventory by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | Yes | Agents, tools, MCP servers, data access and execution paths are inventoried. | [aim-security-54](/vendors/aim-security/#fact-aim-security-54), [aim-security-11](/vendors/aim-security/#fact-aim-security-11) | | [Bay](/vendors/bay/) | Yes | Inventory of agents, MCP servers, extensions and skills across surfaces. | [bay-11](/vendors/bay/#fact-bay-11), [bay-14](/vendors/bay/#fact-bay-14), [bay-37](/vendors/bay/#fact-bay-37) | | [Check Point](/vendors/check-point/) | Yes | Continuously updated inventory of agents, tools and MCP servers. | [check-point-15](/vendors/check-point/#fact-check-point-15), [check-point-16](/vendors/check-point/#fact-check-point-16) | | [CrowdStrike](/vendors/crowdstrike/) | Yes | Discovers agents, MCP servers, skills and tools on endpoints and registers agents in a directory. | [crowdstrike-01](/vendors/crowdstrike/#fact-crowdstrike-01), [crowdstrike-02](/vendors/crowdstrike/#fact-crowdstrike-02), [crowdstrike-32](/vendors/crowdstrike/#fact-crowdstrike-32) | | [Island](/vendors/island/) | Yes | Live inventory of AI assets and non-human identities. | [island-04](/vendors/island/#fact-island-04), [island-12](/vendors/island/#fact-island-12) | | [Lasso Security](/vendors/lasso-security/) | Yes | Agent and tool inventory with a dependency graph. | [lasso-security-32](/vendors/lasso-security/#fact-lasso-security-32), [lasso-security-39](/vendors/lasso-security/#fact-lasso-security-39), [lasso-security-19](/vendors/lasso-security/#fact-lasso-security-19) | | [Microsoft](/vendors/microsoft/) | Yes | Agent registry and AI agent inventory. | [microsoft-31](/vendors/microsoft/#fact-microsoft-31), [microsoft-36](/vendors/microsoft/#fact-microsoft-36) | | [Noma Security](/vendors/noma-security/) | Yes | Continuous registry of agents, MCP servers, skills and models. | [noma-security-25](/vendors/noma-security/#fact-noma-security-25), [noma-security-30](/vendors/noma-security/#fact-noma-security-30) | | [Operant AI](/vendors/operant-ai/) | Yes | Live inventory of agents, tools, MCP servers and skills. | [operant-ai-14](/vendors/operant-ai/#fact-operant-ai-14), [operant-ai-39](/vendors/operant-ai/#fact-operant-ai-39) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | Agent and cloud resource inventory documented. | [palo-alto-networks-35](/vendors/palo-alto-networks/#fact-palo-alto-networks-35), [palo-alto-networks-36](/vendors/palo-alto-networks/#fact-palo-alto-networks-36), [palo-alto-networks-37](/vendors/palo-alto-networks/#fact-palo-alto-networks-37) | | [Pillar Security](/vendors/pillar-security/) | Yes | Agent, MCP server, skill and plugin inventory. | [pillar-security-04](/vendors/pillar-security/#fact-pillar-security-04), [pillar-security-05](/vendors/pillar-security/#fact-pillar-security-05) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Yes | Discovers every agent and MCP server and maps relationships. | [sentinelone-10](/vendors/sentinelone/#fact-sentinelone-10) | | [Straiker](/vendors/straiker/) | Yes | Real-time inventory of agents, tools, MCP servers and Skills. | [straiker-06](/vendors/straiker/#fact-straiker-06), [straiker-26](/vendors/straiker/#fact-straiker-26) | | [WitnessAI](/vendors/witnessai/) | Yes | Inventory of agents, MCP servers and tools they reach. | [witnessai-37](/vendors/witnessai/#fact-witnessai-37), [witnessai-26](/vendors/witnessai/#fact-witnessai-26) | | [Zenity](/vendors/zenity/) | Yes | Live inventory of agents across SaaS, cloud and endpoints. | [zenity-43](/vendors/zenity/#fact-zenity-43), [zenity-32](/vendors/zenity/#fact-zenity-32), [zenity-42](/vendors/zenity/#fact-zenity-42) | | [Cisco](/vendors/cisco/) | Partial | Inventory of AI assets documented; agent and MCP registry in development. | [cisco-42](/vendors/cisco/#fact-cisco-42), [cisco-43](/vendors/cisco/#fact-cisco-43) | ## What is agent inventory as a control point? Agent inventory is the control that discovers the agents, tools and skills running in an organization and records an owner and status for each. It is the first control point because every other control needs a list of what to apply to. Sources: [AI agent support matrix, Microsoft Defender for Endpoint](https://learn.microsoft.com/en-us/defender-endpoint/ai-agent-support-matrix). ## Why it matters and where it stops Microsoft says it expanded discovery by 35 types to 60 supported local agents, including Claude Code, OpenAI Codex and GitHub Copilot, feeding a unified registry. Noma says it connects through the EDR or MDM already in place to turn local evidence into a live inventory across managed devices. An inventory shows what exists, not what it does. It also depends on the layers it can reach: a list built from endpoints will miss agents hosted in SaaS and cloud platforms unless other sources feed it. Sources: [Agents are your newest insiders. Security must move at their speed.](https://www.microsoft.com/en-us/security/blog/2026/10/19/agents-are-your-newest-insiders-security-must-move-at-their-speed/); [Noma Extends Agent Security and Governance to the Employee Endpoint](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html). ## Who documents it Yes on 16 Nov 2026: Aim Security, Bay, Check Point, CrowdStrike, Island, Lasso Security, Microsoft, Noma Security, Operant AI, Palo Alto Networks, Pillar Security, SentinelOne (Prompt Security), Straiker, WitnessAI, Zenity. Partial: Cisco. None found: none. ## Related pages - [The full Agent Control Point Map](/agent-control-point-map/) - [Audit trail](/agent-control-point-map/audit-trail/) - [Posture assessment](/agent-control-point-map/posture-assessment/) - [Method](/agent-control-point-map/methodology/) - [Vendor profiles](/vendors/) --- # Agent posture assessment as an AI agent control point URL: https://aiagentsecuritysolutions.com/agent-control-point-map/posture-assessment/ Updated: 2026-11-16 Posture assessment is the control that checks agents for misconfiguration and risk before they act. On 16 Nov 2026, 11 of 16 vendors had a yes for this control point, 5 had partial and 0 had none found. Levels are from vendor documentation read that day, not lab tests. ## Posture assessment: every vendor **Posture assessment by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | Yes | Risky agent settings and over-privileged access are surfaced. | [aim-security-27](/vendors/aim-security/#fact-aim-security-27), [aim-security-47](/vendors/aim-security/#fact-aim-security-47) | | [Bay](/vendors/bay/) | Yes | Risky configurations and toxic combinations flagged with severity and fix. | [bay-66](/vendors/bay/#fact-bay-66), [bay-67](/vendors/bay/#fact-bay-67) | | [Check Point](/vendors/check-point/) | Yes | Per-agent risk rating with contributing factors. | [check-point-21](/vendors/check-point/#fact-check-point-21), [check-point-18](/vendors/check-point/#fact-check-point-18) | | [Island](/vendors/island/) | Yes | Risk-threshold policy on skills and MCP servers; per-asset scoring is described on the posture page. | [island-06](/vendors/island/#fact-island-06) | | [Lasso Security](/vendors/lasso-security/) | Yes | AI-SPM policies for agent behavior, tools and resources. | [lasso-security-54](/vendors/lasso-security/#fact-lasso-security-54), [lasso-security-39](/vendors/lasso-security/#fact-lasso-security-39) | | [Microsoft](/vendors/microsoft/) | Yes | Posture management with advanced hunting queries for misconfigurations. | [microsoft-29](/vendors/microsoft/#fact-microsoft-29), [microsoft-32](/vendors/microsoft/#fact-microsoft-32) | | [Noma Security](/vendors/noma-security/) | Yes | Posture checks for misconfigurations and toxic combinations. | [noma-security-47](/vendors/noma-security/#fact-noma-security-47), [noma-security-28](/vendors/noma-security/#fact-noma-security-28) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | SaaS Agent Security covers agent configuration and posture risk. | [palo-alto-networks-33](/vendors/palo-alto-networks/#fact-palo-alto-networks-33), [palo-alto-networks-36](/vendors/palo-alto-networks/#fact-palo-alto-networks-36) | | [Pillar Security](/vendors/pillar-security/) | Yes | Configuration audit with SAIL-mapped issues. | [pillar-security-08](/vendors/pillar-security/#fact-pillar-security-08), [pillar-security-16](/vendors/pillar-security/#fact-pillar-security-16) | | [Straiker](/vendors/straiker/) | Yes | Agent-SPM evaluates configurations, permissions and blast radius. | [straiker-16](/vendors/straiker/#fact-straiker-16) | | [Zenity](/vendors/zenity/) | Yes | AISPM evaluates agent configuration and permissions before go-live. | [zenity-44](/vendors/zenity/#fact-zenity-44) | | [Cisco](/vendors/cisco/) | Partial | Page cites detecting misconfigurations and vulnerabilities in AI applications, plus red teaming. | [cisco-02](/vendors/cisco/#fact-cisco-02), [cisco-43](/vendors/cisco/#fact-cisco-43) | | [CrowdStrike](/vendors/crowdstrike/) | Partial | Shows where unmanaged AI risk exists; vulnerability testing of agents via Dreadnode is a planned acquisition, not shipped. | [crowdstrike-01](/vendors/crowdstrike/#fact-crowdstrike-01), [crowdstrike-07](/vendors/crowdstrike/#fact-crowdstrike-07) | | [Operant AI](/vendors/operant-ai/) | Partial | Risk score per tool and supply-chain trust scores; no misconfiguration assessment found. | [operant-ai-39](/vendors/operant-ai/#fact-operant-ai-39) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Partial | Scores agents and MCP servers for risk against policy. | [sentinelone-45](/vendors/sentinelone/#fact-sentinelone-45) | | [WitnessAI](/vendors/witnessai/) | Partial | MCP Catalog scores tools against OWASP and CVE risk classes. | [witnessai-17](/vendors/witnessai/#fact-witnessai-17) | ## What is posture assessment as a control point? Posture assessment examines an agent's configuration, permissions and connected tools for weaknesses, without waiting for an incident. Typical findings are secrets in instructions, unsandboxed execution, and permissions wider than the task needs. Sources: [Noma Extends Agent Security and Governance to the Employee Endpoint](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html). ## Why it matters and where it stops Noma says it continuously assesses discovered assets for risks such as secrets in agent instructions, unsandboxed execution and excessive agency. Zenity Labs' AgentCore research found a default role shared by every agent in an account and region, the kind of finding a posture check on roles is meant to catch. A posture check is a point-in-time or periodic view. It does not stop a prompt-injected agent from misusing access it legitimately holds, which is why it pairs with a runtime control. Sources: [Noma Extends Agent Security and Governance to the Employee Endpoint](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html); [Zenity Labs reports AgentCorruption](https://zenity.io/press-release/zenity-labs-discloses-agentcorruption-a-chain-of-aws-agentcore-flaws). ## Who documents it Yes on 16 Nov 2026: Aim Security, Bay, Check Point, Island, Lasso Security, Microsoft, Noma Security, Palo Alto Networks, Pillar Security, Straiker, Zenity. Partial: Cisco, CrowdStrike, Operant AI, SentinelOne (Prompt Security), WitnessAI. None found: none. ## Related pages - [The full Agent Control Point Map](/agent-control-point-map/) - [Agent inventory](/agent-control-point-map/agent-inventory/) - [Runtime block](/agent-control-point-map/runtime-block/) - [Method](/agent-control-point-map/methodology/) - [Vendor profiles](/vendors/) --- # Runtime block as an AI agent control point URL: https://aiagentsecuritysolutions.com/agent-control-point-map/runtime-block/ Updated: 2026-11-16 Runtime block is the control that stops an agent action before it completes. On 16 Nov 2026, 16 of 16 vendors had a yes for this control point, 0 had partial and 0 had none found. Levels are from vendor documentation read that day, not lab tests. ## Runtime block: every vendor **Runtime block by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | Yes | Risky tool calls and actions are blocked before they execute. | [aim-security-51](/vendors/aim-security/#fact-aim-security-51), [aim-security-52](/vendors/aim-security/#fact-aim-security-52), [aim-security-41](/vendors/aim-security/#fact-aim-security-41) | | [Bay](/vendors/bay/) | Yes | Allow, ask, deny or steer before an action completes. | [bay-47](/vendors/bay/#fact-bay-47), [bay-48](/vendors/bay/#fact-bay-48), [bay-55](/vendors/bay/#fact-bay-55) | | [Check Point](/vendors/check-point/) | Yes | Tool calls evaluated and blocked at runtime via the Guard API. | [check-point-25](/vendors/check-point/#fact-check-point-25), [check-point-26](/vendors/check-point/#fact-check-point-26) | | [Cisco](/vendors/cisco/) | Yes | Blocks prompts, shell commands and tool calls before they run. | [cisco-10](/vendors/cisco/#fact-cisco-10), [cisco-05](/vendors/cisco/#fact-cisco-05), [cisco-02](/vendors/cisco/#fact-cisco-02) | | [CrowdStrike](/vendors/crowdstrike/) | Yes | Hooks block malicious activity before it runs; Kill Switch blocks model access at gateways. | [crowdstrike-10](/vendors/crowdstrike/#fact-crowdstrike-10), [crowdstrike-11](/vendors/crowdstrike/#fact-crowdstrike-11), [crowdstrike-24](/vendors/crowdstrike/#fact-crowdstrike-24) | | [Island](/vendors/island/) | Yes | Inline control points can stop an action before it happens. | [island-16](/vendors/island/#fact-island-16), [island-17](/vendors/island/#fact-island-17) | | [Lasso Security](/vendors/lasso-security/) | Yes | Says it blocks or quarantines actions before they execute. | [lasso-security-10](/vendors/lasso-security/#fact-lasso-security-10), [lasso-security-11](/vendors/lasso-security/#fact-lasso-security-11), [lasso-security-12](/vendors/lasso-security/#fact-lasso-security-12) | | [Microsoft](/vendors/microsoft/) | Yes | Custom rules block tool calls before execution; endpoint runtime block in preview. | [microsoft-28](/vendors/microsoft/#fact-microsoft-28), [microsoft-19](/vendors/microsoft/#fact-microsoft-19) | | [Noma Security](/vendors/noma-security/) | Yes | Blocks actions per policy; options depend on integration (a20). | [noma-security-20](/vendors/noma-security/#fact-noma-security-20), [noma-security-23](/vendors/noma-security/#fact-noma-security-23) | | [Operant AI](/vendors/operant-ai/) | Yes | Inline blocking of out-of-scope and harmful actions before execution. | [operant-ai-23](/vendors/operant-ai/#fact-operant-ai-23), [operant-ai-24](/vendors/operant-ai/#fact-operant-ai-24), [operant-ai-30](/vendors/operant-ai/#fact-operant-ai-30) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | Block actions stop requests; SaaS agents can be paused or terminated. | [palo-alto-networks-34](/vendors/palo-alto-networks/#fact-palo-alto-networks-34), [palo-alto-networks-47](/vendors/palo-alto-networks/#fact-palo-alto-networks-47), [palo-alto-networks-48](/vendors/palo-alto-networks/#fact-palo-alto-networks-48) | | [Pillar Security](/vendors/pillar-security/) | Yes | Pre-tool hook blocking for Claude Code, Codex CLI and Cursor. | [pillar-security-09](/vendors/pillar-security/#fact-pillar-security-09), [pillar-security-10](/vendors/pillar-security/#fact-pillar-security-10) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Yes | States it stops violations before they propagate and enforces at the point of interaction. | [sentinelone-43](/vendors/sentinelone/#fact-sentinelone-43), [sentinelone-41](/vendors/sentinelone/#fact-sentinelone-41) | | [Straiker](/vendors/straiker/) | Yes | Blocks named actions at supported execution points. | [straiker-22](/vendors/straiker/#fact-straiker-22), [straiker-23](/vendors/straiker/#fact-straiker-23) | | [WitnessAI](/vendors/witnessai/) | Yes | Denies off-list tool calls before they execute and blocks prompt injection before agents act. | [witnessai-14](/vendors/witnessai/#fact-witnessai-14), [witnessai-35](/vendors/witnessai/#fact-witnessai-35) | | [Zenity](/vendors/zenity/) | Yes | Boundaries and AIDR block actions before they complete. | [zenity-45](/vendors/zenity/#fact-zenity-45), [zenity-46](/vendors/zenity/#fact-zenity-46), [zenity-52](/vendors/zenity/#fact-zenity-52) | ## What is runtime block as a control point? A runtime block evaluates a proposed action, such as a tool call, a file write or an outbound request, and stops it before it completes. It differs from alerting, which reports an action after it happened. It needs a hook where the action can be intercepted. Sources: [Protect enterprise coding agents with Cisco DefenseClaw and NVIDIA OpenShell](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell). ## Why it matters and where it stops Cisco says DefenseClaw checks every prompt, tool call and tool result inside an OpenShell sandbox and decides each web destination by host name. Operant says its Agent Protector can redact a field from a tool result, block a single outbound call or stop one write while the rest of the task completes. A block only covers the agents and paths that carry the hook. Operant says its Purview path for Copilot detects and alerts but does not sit inline, so the same vendor can have both a blocking and a detect-only route. Sources: [Protect enterprise coding agents with Cisco DefenseClaw and NVIDIA OpenShell](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell); [Operant: coverage for Microsoft Copilot and embedded AI assistants](https://www.operant.ai/art-kubed/announcing-operants-coverage-for-microsoft-copilot-and-every-ai-assistant-embedded-in-microsoft-365). ## Who documents it Yes on 16 Nov 2026: Aim Security, Bay, Check Point, Cisco, CrowdStrike, Island, Lasso Security, Microsoft, Noma Security, Operant AI, Palo Alto Networks, Pillar Security, SentinelOne (Prompt Security), Straiker, WitnessAI, Zenity. Partial: none. None found: none. ## Related pages - [The full Agent Control Point Map](/agent-control-point-map/) - [Posture assessment](/agent-control-point-map/posture-assessment/) - [Permissions and revocation](/agent-control-point-map/permissions-and-revocation/) - [Method](/agent-control-point-map/methodology/) - [Vendor profiles](/vendors/) --- # Permissions and revocation as an AI agent control point URL: https://aiagentsecuritysolutions.com/agent-control-point-map/permissions-and-revocation/ Updated: 2026-11-16 Permissions and revocation is the control that limits what an agent may access and removes that access. On 16 Nov 2026, 5 of 16 vendors had a yes for this control point, 11 had partial and 0 had none found. Levels are from vendor documentation read that day, not lab tests. ## Permissions and revocation: every vendor **Permissions and revocation by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [CrowdStrike](/vendors/crowdstrike/) | Yes | Short-lived least-privilege access brokering and revocation when conditions change. | [crowdstrike-31](/vendors/crowdstrike/#fact-crowdstrike-31), [crowdstrike-33](/vendors/crowdstrike/#fact-crowdstrike-33) | | [Island](/vendors/island/) | Yes | Per tool, per action and per actor limits through the MCP Gateway; just-in-time credentials. | [island-13](/vendors/island/#fact-island-13), [island-15](/vendors/island/#fact-island-15) | | [Microsoft](/vendors/microsoft/) | Yes | Conditional Access and Dynamic Permissions for agents. | [microsoft-34](/vendors/microsoft/#fact-microsoft-34), [microsoft-55](/vendors/microsoft/#fact-microsoft-55) | | [Noma Security](/vendors/noma-security/) | Yes | Tool and action level policy by identity. | [noma-security-23](/vendors/noma-security/#fact-noma-security-23), [noma-security-17](/vendors/noma-security/#fact-noma-security-17), [noma-security-37](/vendors/noma-security/#fact-noma-security-37) | | [Operant AI](/vendors/operant-ai/) | Yes | Agent access enforcement and least-privilege MCP tool use. | [operant-ai-19](/vendors/operant-ai/#fact-operant-ai-19), [operant-ai-21](/vendors/operant-ai/#fact-operant-ai-21) | | [Aim Security](/vendors/aim-security/) | Partial | Over-privilege is spotted; revocation is not documented. | [aim-security-47](/vendors/aim-security/#fact-aim-security-47), [aim-security-48](/vendors/aim-security/#fact-aim-security-48) | | [Bay](/vendors/bay/) | Partial | Policy scoping and lock-down of reachable credentials; least-privilege revocation not specifically documented. | [bay-31](/vendors/bay/#fact-bay-31), [bay-41](/vendors/bay/#fact-bay-41), [bay-42](/vendors/bay/#fact-bay-42) | | [Check Point](/vendors/check-point/) | Partial | Tool and MCP allow and deny lists; no least-privilege or revocation of agent credentials found. | [check-point-27](/vendors/check-point/#fact-check-point-27), [check-point-21](/vendors/check-point/#fact-check-point-21) | | [Cisco](/vendors/cisco/) | Partial | Group-based policy and MCP server blocking; Duo IAM agent permissions planned. | [cisco-16](/vendors/cisco/#fact-cisco-16), [cisco-31](/vendors/cisco/#fact-cisco-31), [cisco-40](/vendors/cisco/#fact-cisco-40) | | [Lasso Security](/vendors/lasso-security/) | Partial | Role-based permissions and overprivileged-agent flagging; revocation not documented. | [lasso-security-24](/vendors/lasso-security/#fact-lasso-security-24), [lasso-security-34](/vendors/lasso-security/#fact-lasso-security-34) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Partial | Identity governance for agents is documented; least-privilege and revocation details were not read. | [palo-alto-networks-39](/vendors/palo-alto-networks/#fact-palo-alto-networks-39), [palo-alto-networks-40](/vendors/palo-alto-networks/#fact-palo-alto-networks-40) | | [Pillar Security](/vendors/pillar-security/) | Partial | Tool allow-lists per MCP server and RBAC for AI assets; no revocation found. | [pillar-security-14](/vendors/pillar-security/#fact-pillar-security-14), [pillar-security-23](/vendors/pillar-security/#fact-pillar-security-23) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Partial | Least-privilege enforcement stated in general terms. | [sentinelone-38](/vendors/sentinelone/#fact-sentinelone-38) | | [Straiker](/vendors/straiker/) | Partial | Overprivilege is identified and tool access can be revoked via the kill switch. | [straiker-15](/vendors/straiker/#fact-straiker-15), [straiker-17](/vendors/straiker/#fact-straiker-17) | | [WitnessAI](/vendors/witnessai/) | Partial | Organization-wide approved tool list; no credential least-privilege or revocation found. | [witnessai-34](/vendors/witnessai/#fact-witnessai-34), [witnessai-32](/vendors/witnessai/#fact-witnessai-32) | | [Zenity](/vendors/zenity/) | Partial | Identity-aware policy and privilege visibility; revocation workflow not specifically documented in public pages. | [zenity-35](/vendors/zenity/#fact-zenity-35), [zenity-36](/vendors/zenity/#fact-zenity-36) | ## What is permissions and revocation as a control point? This control point sets least privilege for an agent and takes access back when the task ends or risk rises. It covers scopes on tokens, role ceilings and time limits on grants. AWS describes a permissions boundary as setting the maximum permissions an identity-based policy can grant to an IAM entity. Sources: [Agents are your newest insiders. Security must move at their speed.](https://www.microsoft.com/en-us/security/blog/2026/10/19/agents-are-your-newest-insiders-security-must-move-at-their-speed/). ## Why it matters and where it stops Microsoft says Dynamic Permissions let organization-wide policy set the outer boundary and scope authority to the task, with time limits and removal as the task changes. CrowdStrike says its identity provider brokers short-lived, tightly scoped access in place of standing credentials. Permissions limit what an agent can reach, not what it chooses to do with that reach. They also depend on the agent using the brokered path: a credential stored in the agent's environment can bypass a broker. Sources: [CrowdStrike Announces Agentic Identity Provider](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/); [Permissions boundaries for IAM entities](https://docs.aws.amazon.com/IAM/latest/UserGuide/access_policies_boundaries.html). ## Who documents it Yes on 16 Nov 2026: CrowdStrike, Island, Microsoft, Noma Security, Operant AI. Partial: Aim Security, Bay, Check Point, Cisco, Lasso Security, Palo Alto Networks, Pillar Security, SentinelOne (Prompt Security), Straiker, WitnessAI, Zenity. None found: none. ## Related pages - [The full Agent Control Point Map](/agent-control-point-map/) - [Runtime block](/agent-control-point-map/runtime-block/) - [Audit trail](/agent-control-point-map/audit-trail/) - [Method](/agent-control-point-map/methodology/) - [Vendor profiles](/vendors/) --- # Agent audit trail as an AI agent control point URL: https://aiagentsecuritysolutions.com/agent-control-point-map/audit-trail/ Updated: 2026-11-16 An audit trail is the control that records what an agent did and what each control decided. On 16 Nov 2026, 15 of 16 vendors had a yes for this control point, 1 had partial and 0 had none found. Levels are from vendor documentation read that day, not lab tests. ## Audit trail: every vendor **Audit trail by vendor** | Vendor | Level | Note | Evidence | |---|---|---|---| | [Aim Security](/vendors/aim-security/) | Yes | Every interaction is logged in one data lake and dashboard. | [aim-security-52](/vendors/aim-security/#fact-aim-security-52), [aim-security-56](/vendors/aim-security/#fact-aim-security-56), [aim-security-68](/vendors/aim-security/#fact-aim-security-68) | | [Bay](/vendors/bay/) | Yes | Decision log, session timelines and SIEM export in OCSF. | [bay-61](/vendors/bay/#fact-bay-61), [bay-62](/vendors/bay/#fact-bay-62), [bay-63](/vendors/bay/#fact-bay-63) | | [Cisco](/vendors/cisco/) | Yes | Block and approval records exported to Splunk or OpenTelemetry. | [cisco-17](/vendors/cisco/#fact-cisco-17), [cisco-18](/vendors/cisco/#fact-cisco-18), [cisco-35](/vendors/cisco/#fact-cisco-35) | | [CrowdStrike](/vendors/crowdstrike/) | Yes | Agent sessions, tool calls and MCP servers linked to execution, with SIEM export. | [crowdstrike-38](/vendors/crowdstrike/#fact-crowdstrike-38), [crowdstrike-40](/vendors/crowdstrike/#fact-crowdstrike-40), [crowdstrike-41](/vendors/crowdstrike/#fact-crowdstrike-41) | | [Island](/vendors/island/) | Yes | Log entry on every intervention; human versus agent recorded in audit logs. | [island-19](/vendors/island/#fact-island-19), [island-25](/vendors/island/#fact-island-25) | | [Lasso Security](/vendors/lasso-security/) | Yes | Audit trails of connections, tool calls and agent decisions, exportable. | [lasso-security-25](/vendors/lasso-security/#fact-lasso-security-25), [lasso-security-40](/vendors/lasso-security/#fact-lasso-security-40) | | [Microsoft](/vendors/microsoft/) | Yes | Purview keeps agent activity records. | [microsoft-37](/vendors/microsoft/#fact-microsoft-37) | | [Noma Security](/vendors/noma-security/) | Yes | Observability of prompts and tool calls plus audit trails. | [noma-security-31](/vendors/noma-security/#fact-noma-security-31), [noma-security-48](/vendors/noma-security/#fact-noma-security-48) | | [Operant AI](/vendors/operant-ai/) | Yes | Traces, audit logs and SIEM export. | [operant-ai-37](/vendors/operant-ai/#fact-operant-ai-37), [operant-ai-38](/vendors/operant-ai/#fact-operant-ai-38), [operant-ai-40](/vendors/operant-ai/#fact-operant-ai-40) | | [Palo Alto Networks](/vendors/palo-alto-networks/) | Yes | Auditability and logs of tool invocations and verdicts documented. | [palo-alto-networks-36](/vendors/palo-alto-networks/#fact-palo-alto-networks-36), [palo-alto-networks-46](/vendors/palo-alto-networks/#fact-palo-alto-networks-46) | | [Pillar Security](/vendors/pillar-security/) | Yes | Activity timeline and audit trail. | [pillar-security-12](/vendors/pillar-security/#fact-pillar-security-12), [pillar-security-24](/vendors/pillar-security/#fact-pillar-security-24) | | [SentinelOne (Prompt Security)](/vendors/sentinelone/) | Yes | Searchable log of every session, action and tool. | [sentinelone-44](/vendors/sentinelone/#fact-sentinelone-44) | | [Straiker](/vendors/straiker/) | Yes | Audit logs, dashboards and preserved activity trail. | [straiker-27](/vendors/straiker/#fact-straiker-27), [straiker-24](/vendors/straiker/#fact-straiker-24) | | [WitnessAI](/vendors/witnessai/) | Yes | Every blocked tool call records user, agent, tool and rule. | [witnessai-40](/vendors/witnessai/#fact-witnessai-40) | | [Zenity](/vendors/zenity/) | Yes | Step-level activity table and audit trails documented. | [zenity-27](/vendors/zenity/#fact-zenity-27), [zenity-54](/vendors/zenity/#fact-zenity-54), [zenity-28](/vendors/zenity/#fact-zenity-28) | | [Check Point](/vendors/check-point/) | Partial | Logs API and analytics exist; audit trail is described in general terms. | [check-point-36](/vendors/check-point/#fact-check-point-36), [check-point-33](/vendors/check-point/#fact-check-point-33) | ## What is audit trail as a control point? An audit trail records agent sessions, tool calls, blocked and allowed actions, and the policy behind each decision, in a form that can go to a SIEM. It supports investigation and compliance and shows whether other controls worked. Sources: [Protect enterprise coding agents with Cisco DefenseClaw and NVIDIA OpenShell](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell). ## Why it matters and where it stops Cisco says every block, approval and decision is recorded with the agent, sandbox and policy behind it and can be sent to Splunk or any OpenTelemetry backend. NIST, the NSA and CISA list detection and reporting among five incident stages in ten key practices published on 16 October 2026, which depend on such records. A trail records what the control saw. Activity that never passed through a monitored layer leaves no entry, and a record does not stop an action. Sources: [Protect enterprise coding agents with Cisco DefenseClaw and NVIDIA OpenShell](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell); [Key Practices for Incident Readiness and Response](https://www.nist.gov/news-events/news/2026/10/key-practices-incident-readiness-and-response-super-intelligence-training). ## Who documents it Yes on 16 Nov 2026: Aim Security, Bay, Cisco, CrowdStrike, Island, Lasso Security, Microsoft, Noma Security, Operant AI, Palo Alto Networks, Pillar Security, SentinelOne (Prompt Security), Straiker, WitnessAI, Zenity. Partial: Check Point. None found: none. ## Related pages - [The full Agent Control Point Map](/agent-control-point-map/) - [Permissions and revocation](/agent-control-point-map/permissions-and-revocation/) - [Agent inventory](/agent-control-point-map/agent-inventory/) - [Method](/agent-control-point-map/methodology/) - [Vendor profiles](/vendors/) --- # AI agent security datasets URL: https://aiagentsecuritysolutions.com/datasets/ Updated: 2026-11-16 Two datasets are free to download as CSV and JSON under CC BY 4.0: the Layer Coverage Map, which places sixteen vendors across five layers, and the Agent Control Point Map, which shows where in an agent's run each vendor acts. Every cell cites fact ids from the 16 Nov 2026 evidence files. ## Downloads - [Layer Coverage Map](/datasets/layer-coverage-map/): [CSV](/datasets/layer-coverage-map.csv), [JSON](/datasets/layer-coverage-map.json) - [Agent Control Point Map](/datasets/agent-control-point-map/): [CSV](/datasets/agent-control-point-map.csv), [JSON](/datasets/agent-control-point-map.json) ## License and citation CC BY 4.0. AI Agent Security Solutions, edition 16 Nov 2026, https://aiagentsecuritysolutions.com/, CC BY 4.0. ## Related pages - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) - [Method](/method/) - [The ranking](/ranking/) --- # Layer Coverage Map dataset: CSV and JSON download URL: https://aiagentsecuritysolutions.com/datasets/layer-coverage-map/ Updated: 2026-11-16 The Layer Coverage Map is published as a CSV file and a JSON file under CC BY 4.0. It holds 80 vendor and dimension cells across 5 dimensions, read on 2026-11-16. Cite it with the line below when you reuse the data, and link back to this page so readers can check the files against the sources. ## Files - [Layer Coverage Map as CSV](/datasets/layer-coverage-map.csv) - [Layer Coverage Map as JSON](/datasets/layer-coverage-map.json) ## How to cite AI Agent Security Solutions, edition 16 Nov 2026, https://aiagentsecuritysolutions.com/, CC BY 4.0. ## Related pages - [The Layer Coverage Map](/layer-coverage-map/) - [Method](/layer-coverage-map/methodology/) - [All datasets](/datasets/) - [Scoring method](/method/) --- # Agent Control Point Map dataset: CSV and JSON download URL: https://aiagentsecuritysolutions.com/datasets/agent-control-point-map/ Updated: 2026-11-16 The Agent Control Point Map is published as a CSV file and a JSON file under CC BY 4.0. It holds 80 vendor and dimension cells across 5 dimensions, read on 2026-11-16. Cite it with the line below when you reuse the data, and link back to this page so readers can check the files against the sources. ## Files - [Agent Control Point Map as CSV](/datasets/agent-control-point-map.csv) - [Agent Control Point Map as JSON](/datasets/agent-control-point-map.json) ## How to cite AI Agent Security Solutions, edition 16 Nov 2026, https://aiagentsecuritysolutions.com/, CC BY 4.0. ## Related pages - [The Agent Control Point Map](/agent-control-point-map/) - [Method](/agent-control-point-map/methodology/) - [All datasets](/datasets/) - [Scoring method](/method/) --- # AI agent security guides URL: https://aiagentsecuritysolutions.com/guides/ Updated: 2026-11-16 Twelve guides cover securing AI agents by layer: the five layers, risks, best practices, platform or point solutions, frameworks, metrics, incident response, multi-agent systems, red team exercises, small teams, a market map and who owns each layer. Each cites vendor documents and primary sources read on 16 Nov 2026. ## 12 guides - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/): AI agent security best practices for each layer, from least privilege and hooks to scoped tokens, gateways and logs, led by those that stop actions. - [AI Agent Security for Multi-Agent Systems](/guides/ai-agent-security-for-multi-agent-systems/): How multi-agent systems change AI agent security: trust between agents, delegated credentials, tracing agent chains and containing a compromised peer. - [AI Agent Security for Small Teams: Which Layer First](/guides/ai-agent-security-for-small-teams-which-layer-first/): A small team's plan for AI agent security: which layer to cover first, what can wait, and low-cost options from platform tools and open-source projects. - [AI Agent Security Frameworks Mapped to Five Layers](/guides/ai-agent-security-frameworks-mapped-to-layers/): How OWASP, NIST and MITRE ATLAS guidance on AI agents maps to the endpoint, browser, cloud workspace, identity and gateway layers. - [AI Agent Security Incident Response Playbook](/guides/ai-agent-security-incident-response-playbook/): A step-by-step playbook for when an AI agent misbehaves: contain, revoke credentials, review the trail, find the cause and restore with tighter limits. - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/): The five layers of AI agent security, endpoint, browser, cloud workspace, identity and gateway, what each can see and block, and which to cover first. - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/): Which metrics to track for AI agent security, from agent inventory coverage to blocked actions and time to revoke, and what a monthly report should hold. - [AI Agent Security Platform or Point Solutions?](/guides/ai-agent-security-platform-or-point-solutions/): How to choose between one AI agent security platform and several point solutions: what a platform is, its layer coverage and when one product is enough. - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/): The main AI agent security risks, from prompt injection to excessive agency and runaway use, and the layer best placed to stop each one. - [AI Agent Security Roles: Who Owns Each Layer](/guides/ai-agent-security-roles-who-owns-each-layer/): Who owns each layer of AI agent security: endpoint, browser, cloud workspace, identity and gateway, and how security, IT and engineering share the work. - [AI Agent Security Solutions Market Map, November 2026](/guides/ai-agent-security-solutions-market-map/): The kinds of vendors selling AI agent security in November 2026: endpoint, network, enterprise browser, agent specialists and cloud platform owners. - [AI Agent Security Testing and Red Team Exercises](/guides/ai-agent-security-testing-and-red-team-exercises/): How to test AI agent security controls and run red team exercises by layer: scope, safe test cases, what to record and how to fix what you find. ## Related pages - [Glossary](/glossary/) - [The ranking](/ranking/) - [FAQ](/faq/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # AI agent security best practices, by layer URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-best-practices/ Updated: 2026-11-16 Good AI agent security practice limits what an agent can do before it acts. Apply least functionality, least permission and limited autonomy at each layer, put a blocking control on the tool call, issue short-lived scoped credentials, route traffic through a control point and keep logs that name the agent, the user and the tool. ## What are the best practices for AI agent security? Cut the three root causes OWASP names for excessive agency: excessive functionality, excessive permissions and excessive autonomy. Give each agent only the tools it needs, only the permissions those tools need, and a human approval step for actions that are hard to undo. Apply the same rule at every layer rather than once in the model prompt, because a prompt is advice to the model and a control is enforced outside it. Start with an inventory, since you cannot limit what you have not found. List agents, the tools each can call, the credentials each holds and the data each can reach. Then apply the three limits to the list, agent by agent, beginning with the agents that can write, send or spend. - Remove tools and plugins the agent does not use. - Scope each tool to read-only where writing is not needed. - Require approval for deletes, payments and external sends. Sources: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/). ## Which practices stop an action rather than record it? Controls that run before the call executes stop an action: a pre-tool hook on the endpoint, an allow list at a gateway, a Conditional Access block on the token, or a guardrail attached to the platform agent. Logging records the action after the fact. Claude Code documents PreToolUse as running before a tool call executes and able to block it. Cisco's DefenseClaw says its guardrail checks a prompt, shell command or tool call before the agent acts. Bay says its enforcer decides allow, ask, deny or steer on every action. Approval steps belong on the actions that are hard to undo: deleting data, sending messages outside the company, changing access and moving money. An ask decision that pauses the agent for a person is a stop in the same sense as a block, and it keeps the agent useful for the rest of its work. Sources: [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks); [Cisco DefenseClaw: guardrail](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/); [Bay: agent runtime enforcement](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide). ## How should you handle credentials and tokens? Issue short-lived, narrowly scoped credentials per task and keep the underlying secret away from the agent. Revoke on a schedule and on any suspicion. RFC 9700 collects current OAuth security practice, and RFC 8693 defines token exchange, including delegation, which lets a service swap one token for another with narrower rights. Island says its MCP Gateway issues just-in-time credentials scoped to one task so the agent never sees the underlying secret. Rotate on a schedule that matches the exposure. An agent that runs for minutes needs a credential that lives for minutes. Keep a list of standing credentials still held by agents and cut it each month. Where a service cannot issue short-lived credentials, place a gateway or broker in front of it. Sources: [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700); [RFC 8693, OAuth 2.0 Token Exchange](https://datatracker.ietf.org/doc/html/rfc8693); [Island: agentic identity](https://www.island.io/product/ai/agentic-identity). ## How do you control which tools and MCP servers an agent can call? Keep an approved list and deny everything else. WitnessAI says a tool or MCP server not on the approved list is denied before the tool executes. The MCP specification treats a protected MCP server as an OAuth 2.1 resource server, so require authorization on every server you approve. Review remote servers before approval and pin versions of local ones. For local MCP servers, record the package name, version and publisher at approval, and compare on every update. A server that changes its own tool descriptions after approval deserves a re-review. Remote servers should authenticate every client and issue tokens scoped to that server. Sources: [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/); [MCP specification: authorization](https://modelcontextprotocol.io/specification/2025-06-18/basic/authorization). ## What should you log for every agent? Log the agent, the user it acts for, the tool, the arguments class, the decision and the rule that made it. Without those five fields an investigation cannot say who or why. OpenTelemetry publishes GenAI semantic conventions, including agent spans, so logs from different tools can share field names. Forward them to the SIEM you already run. Keep logs where an investigator can reach them and where a rogue agent cannot edit them. Send them to a store the agent has no write access to. Set a retention period long enough to cover your slowest detection, commonly several months. Sources: [OpenTelemetry GenAI semantic conventions](https://opentelemetry.io/docs/specs/semconv/gen-ai/). ## How do you test that a control works? Test each control with a harmless action it should block, then confirm the block, the log entry and the alert. Repeat after every agent or policy update. Keep a list of the controls and the date each was last tested, and put it in the monthly report. Include a test of the human path as well: confirm that the person who gets an approval request can read what the agent is asking to do and has the authority to decide. An approval prompt that nobody understands becomes a click-through. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## What belongs in a first-month plan? Week one: inventory agents and approved tools. Week two: deploy one blocking control on the first layer in monitor mode. Week three: move the clear cases to block and scope the credentials of the five most capable agents. Week four: test, log and write the first report. Each step uses controls you can install through tools you already run, so the plan needs process more than new products. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework); [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks). ## Sources - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks) (docs.anthropic.com), read 2026-11-16 - [Cisco DefenseClaw: guardrail](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/) (cisco-ai-defense.github.io), read 2026-11-16 - [Bay: agent runtime enforcement](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide) (bay.io), read 2026-11-16 - [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700) (rfc-editor.org), read 2026-11-16 - [RFC 8693, OAuth 2.0 Token Exchange](https://datatracker.ietf.org/doc/html/rfc8693) (datatracker.ietf.org), read 2026-11-16 - [Island: agentic identity](https://www.island.io/product/ai/agentic-identity) (island.io), read 2026-11-16 - [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/) (witness.ai), read 2026-11-16 - [MCP specification: authorization](https://modelcontextprotocol.io/specification/2025-06-18/basic/authorization) (modelcontextprotocol.io), read 2026-11-16 - [OpenTelemetry GenAI semantic conventions](https://opentelemetry.io/docs/specs/semconv/gen-ai/) (opentelemetry.io), read 2026-11-16 - [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) (nist.gov), read 2026-11-16 ## Related pages - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [AI Agent Security Testing and Red Team Exercises](/guides/ai-agent-security-testing-and-red-team-exercises/) - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [Guardrail policy (glossary)](/glossary/guardrail-policy/) - [Permission boundary (glossary)](/glossary/permission-boundary/) - [All guides](/guides/) - [AI Agent Security for Multi-Agent Systems](/guides/ai-agent-security-for-multi-agent-systems/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security for multi-agent systems URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-for-multi-agent-systems/ Updated: 2026-11-16 In a multi-agent system one agent's output becomes another agent's input, so a compromised or manipulated peer can pass bad instructions down the chain. Give each agent its own identity, narrow its tools, use delegated rather than shared credentials, trace every hop and treat peer messages as untrusted input. ## What changes when agents call agents? Trust now flows between agents, so one weak agent can pull others into harm. OWASP's excessive agency entry lists a malicious or compromised peer agent as a trigger in multi-agent and collaborative systems. Each hop adds permissions that the caller may not have intended. Treat a message from another agent like input from any other untrusted source. Orchestration adds a new role: the agent that assigns work to others. Its credentials and its instructions deserve the strongest protection, because whoever controls the orchestrator controls the workers. Sources: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/). ## How should agents authenticate to each other? Give every agent its own identity and have it act with delegated, scoped tokens, not a shared secret. RFC 8693 defines OAuth token exchange, including delegation, so a service can pass on narrower rights. Microsoft documents agent identity accounts in Entra, and Zenity says it reads agent identities from Entra ID and Okta with their privilege level. Avoid giving worker agents the orchestrator's token. Use delegation so each worker acts with its own narrower token, and so logs show which agent made which call and on whose behalf. Sources: [RFC 8693, OAuth 2.0 Token Exchange](https://datatracker.ietf.org/doc/html/rfc8693); [Microsoft Entra: Conditional Access for agents](https://learn.microsoft.com/en-us/entra/identity/conditional-access/agent-id); [Zenity: agentic identity and access management](https://zenity.io/platform/agentic-identity-and-access-management). ## How do you trace a chain of agents? Carry one trace identifier across every hop and log each tool call with the agent, the caller and the decision. Without it, you see single calls but not the chain. OpenTelemetry's GenAI conventions define agent spans. Operant says it traces agents from prompts through tool calls to memory access, and Check Point says discovery records each agent's owner, model, tools and MCP servers. Record the parent and child relationship for every call. When a user asks why an agent wrote to a file, you should be able to walk back from the write to the user request, through each agent that touched it. Sources: [OpenTelemetry GenAI semantic conventions](https://opentelemetry.io/docs/specs/semconv/gen-ai/); [Operant AI: Agent Protector](https://www.operant.ai/platform/agent-protector); [Check Point: AI agent security](https://www.checkpoint.com/ai-security/ai-agent-security/). ## How do you limit the blast radius? Limit what each agent can reach and what it can ask another agent to do. A planner agent should not hold the credentials its worker agents use. Operant says ScopeGuard blocks out-of-scope actions before execution, limiting the data sources and APIs an agent may use. Set budgets for the chain as well as for each agent: maximum steps, maximum tool calls and maximum spend. A loop between two agents can run for a long time without any single step looking unusual. Sources: [Operant AI: Agent Protector](https://www.operant.ai/platform/agent-protector). ## What about shared memory and context? Shared memory is an attack path. NIST's request for information names risks from models meeting adversarial data, and a shared store lets one poisoned entry reach many agents. Separate memory per agent where you can, and validate what enters shared stores. Treat content that one agent passes to another as data, not as instructions. Strip or flag instruction-like text from tool results before it reaches a planner, and record when you do. Sources: [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems). ## Which layers matter most? Identity and the gateway matter most for multi-agent systems, because that is where cross-agent calls and tokens pass. The cloud workspace layer matters when the agents are built on a SaaS platform. Zenity documents agent identity controls and Palo Alto Networks documents an AI gateway; check the layer pages for the others. Test the chain with a harmless failure: make one worker return a wrong result and watch whether the others notice. The answer shows whether trust between agents is earned or assumed. Sources: [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview); [Zenity: agentic identity and access management](https://zenity.io/platform/agentic-identity-and-access-management). ## Sources - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [RFC 8693, OAuth 2.0 Token Exchange](https://datatracker.ietf.org/doc/html/rfc8693) (datatracker.ietf.org), read 2026-11-16 - [Microsoft Entra: Conditional Access for agents](https://learn.microsoft.com/en-us/entra/identity/conditional-access/agent-id) (learn.microsoft.com), read 2026-11-16 - [Zenity: agentic identity and access management](https://zenity.io/platform/agentic-identity-and-access-management) (zenity.io), read 2026-11-16 - [OpenTelemetry GenAI semantic conventions](https://opentelemetry.io/docs/specs/semconv/gen-ai/) (opentelemetry.io), read 2026-11-16 - [Operant AI: Agent Protector](https://www.operant.ai/platform/agent-protector) (operant.ai), read 2026-11-16 - [Check Point: AI agent security](https://www.checkpoint.com/ai-security/ai-agent-security/) (checkpoint.com), read 2026-11-16 - [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems) (nist.gov), read 2026-11-16 - [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview) (docs.paloaltonetworks.com), read 2026-11-16 ## Related pages - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Incident Response Playbook](/guides/ai-agent-security-incident-response-playbook/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [Multi-agent system (glossary)](/glossary/multi-agent-system/) - [A2A protocol (glossary)](/glossary/a2a-protocol/) - [All guides](/guides/) - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [AI Agent Security for Small Teams: Which Layer First](/guides/ai-agent-security-for-small-teams-which-layer-first/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security for small teams: which layer first URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-for-small-teams-which-layer-first/ Updated: 2026-11-16 A small team should secure the one layer where its agents run most, usually the endpoint for developer-heavy teams or the cloud workspace for SaaS builders, and defer the rest. Use controls already licensed, install through the MDM you run, keep a short approved list and review it monthly. ## Which single layer gives the most cover? For a team of developers, the endpoint gives the most cover because coding agents, local MCP servers and their credentials live there. For a team that builds agents in Copilot Studio or Agentforce, the cloud workspace and identity layers give more. Check where your agents run before you choose. Ask each developer which agents they use, and list the SaaS platforms where staff build agents. A quick test helps: if an agent action on a laptop could expose your source code or cloud keys, the endpoint comes first. If the worst action happens inside a SaaS platform that holds customer data, the cloud workspace comes first. Sources: [What is Amazon Bedrock AgentCore](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html); [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance). ## What can a small team skip? A small team can skip a dedicated gateway until agents call external models and tools from servers. It can also skip a multi-product platform and full red-team exercises in the first quarter. It cannot skip an inventory, an approved list of tools, least privilege and a way to revoke. Those cost little and cover the largest risks OWASP names. Small teams also skip long evaluations. A two-week pilot of one control against your real agents tells you more than a feature matrix. Choose a product that documents deployment through the tools you run and a way out if it does not fit. Sources: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/). ## What low-cost options exist? Start with what you already license and with open-source and free editions. Microsoft lists Agent 365 at $15.00 per user per month, billed yearly, and documents Defender discovery of local agents. Cisco publishes DefenseClaw as Apache-2.0 code and describes an Explorer Edition as free. Claude Code's hooks let you add a pre-tool check yourself. Most commercial startups publish no price, so ask for a quote early. Free does not mean unsupported by evidence. Check that a free edition covers the layer you need and the agents you run, and note the license of any open-source code before you ship it in your tooling. Sources: [Microsoft Agent 365](https://www.microsoft.com/en-us/microsoft-agent-365); [What's new in Microsoft Defender for Endpoint](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint); [Cisco DefenseClaw: guardrail](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/); [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks); [CRN: Cisco unveils zero trust for AI agents](https://www.crn.com/news/security/2026/cisco-unveils-zero-trust-for-ai-agents-5-things-to-know). ## How do you roll out in a week? Pick one control and one group. Push it through the MDM or EDR you run to a pilot group, run it in a monitor mode for a week, then switch the clear cases to block. Vendors publish MDM paths: Pillar Security names Jamf Pro, Kandji and JumpCloud, and Operant names Jamf, Intune, Workspace ONE and Kandji. Measure the pilot: how many agents were found, how many actions were blocked and how many blocks were wrong. A control that blocks wrongly too often will be switched off, so tune before you widen. Sources: [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint); [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector). ## What policy should a small team write first? Write one page: approved agents, approved MCP servers, what data may not be pasted, and who approves exceptions. Keep it short enough that people read it. Review it monthly in the first quarter, then quarterly. Include contractors and personal devices in the policy if they touch your code or data. State what is allowed there, since your MDM may not reach them. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## When should a small team add a second layer? Add the second layer when the first is in block mode and a new agent surface appears, such as staff building agents in a SaaS platform. Add identity controls when agents hold credentials you cannot rotate quickly. Use the layer views on this site to shortlist for the second layer. Revisit the plan every quarter. A small team's agent use changes fast, and a plan that fit in January may miss a platform adopted by March. Sources: [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700). ## Sources - [What is Amazon Bedrock AgentCore](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html) (docs.aws.amazon.com), read 2026-11-16 - [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance) (learn.microsoft.com), read 2026-11-16 - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [Microsoft Agent 365](https://www.microsoft.com/en-us/microsoft-agent-365) (microsoft.com), read 2026-11-16 - [What's new in Microsoft Defender for Endpoint](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint) (learn.microsoft.com), read 2026-11-16 - [Cisco DefenseClaw: guardrail](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/) (cisco-ai-defense.github.io), read 2026-11-16 - [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks) (docs.anthropic.com), read 2026-11-16 - [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint) (pillar.security), read 2026-11-16 - [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector) (operant.ai), read 2026-11-16 - [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) (nist.gov), read 2026-11-16 - [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700) (rfc-editor.org), read 2026-11-16 ## Related pages - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [AI Agent Security Platform or Point Solutions?](/guides/ai-agent-security-platform-or-point-solutions/) - [Blast radius (glossary)](/glossary/blast-radius/) - [All guides](/guides/) - [AI Agent Security for Multi-Agent Systems](/guides/ai-agent-security-for-multi-agent-systems/) - [AI Agent Security Frameworks Mapped to Five Layers](/guides/ai-agent-security-frameworks-mapped-to-layers/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security frameworks mapped to layers URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-frameworks-mapped-to-layers/ Updated: 2026-11-16 OWASP, NIST and MITRE publish guidance that applies to AI agents, and each maps onto the five layers. OWASP names the risks, NIST frames governance and adversarial attacks, MITRE ATLAS catalogs attacker techniques, and the MCP specification and OAuth documents govern the tool and identity layers. Read on 16 Nov 2026. ## Which frameworks apply to AI agents? Five public sources apply: the OWASP Top 10 for Agentic Applications for 2026, the OWASP LLM Top 10, the NIST AI Risk Management Framework with its GenAI profile, MITRE ATLAS, and, for tools, the MCP specification. OWASP published its agentic list on 9 Dec 2025 after work with more than 100 experts. NIST says the AI RMF 1.0 is being revised under the White House AI Action Plan, so check the current version. None of these sources is a certification. They are shared vocabularies that auditors, vendors and regulators recognize, so mapping your controls to them saves time when you are asked to show your work. Keep the mapping as a living table rather than a one-time report. Sources: [OWASP Top 10 for Agentic Applications for 2026](https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/); [OWASP Top 10 for LLM and Gen AI Apps (2025)](https://genai.owasp.org/llm-top-10/); [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework); [NIST AI 600-1, the GenAI Profile](https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf); [MITRE ATLAS](https://atlas.mitre.org/); [MCP specification: security best practices](https://modelcontextprotocol.io/specification/2025-06-18/basic/security_best_practices). ## How do NIST and OWASP guidance map to the five layers? OWASP risks map to layers by where the harmful step shows. Excessive agency maps to identity and endpoint controls, supply chain maps to the endpoint and gateway, and exposure of sensitive information maps to the browser and gateway. NIST's AI RMF is organized around governance and risk management rather than layers, so use it to set the program: owners, risk tolerance and review. Use the layer map to place the controls. A worked example helps. Take excessive agency: the endpoint control is a pre-tool check, the identity control is a narrow scope and short lifetime, the gateway control is an allow list, and the cloud workspace control is the platform's own connector and data policy. One framework item, four layer controls. Sources: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/); [OWASP Top 10 for LLM and Gen AI Apps (2025)](https://genai.owasp.org/llm-top-10/); [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## What does NIST say about securing agent systems? NIST's Center for AI Standards and Innovation issued a request for information on securing AI agent systems in January 2026. It says agent systems can plan and take autonomous actions that affect real-world systems. It names three risk groups: adversarial data such as indirect prompt injection, insecure models such as ones with poisoned data, and harmful actions even without adversarial inputs. Each group maps to a different layer. Read the request for information as a statement of concerns that NIST wants input on, not as a final standard. It signals which risks the U.S. standards body considers distinct for agent systems, which is useful when you decide what to test and what to ask vendors. Sources: [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems). ## How does MITRE ATLAS help? ATLAS gives names to attacker behavior against AI-enabled systems, which helps when you write detections and test plans. On 16 Nov 2026 it listed 16 tactics, 221 techniques and 44 mitigations. Map each technique you care about to the layer where you would see it, then check that a control exists there. ATLAS also publishes mitigations and case studies. A case study shows how a real incident moved through techniques, which is a good starting point for a tabletop exercise with your own agents in the role of the target. Sources: [MITRE ATLAS](https://atlas.mitre.org/); [NIST AI 100-2 E2025, Adversarial Machine Learning](https://csrc.nist.gov/pubs/ai/100/2/e2025/final). ## Which standards govern the identity and tool layers? OAuth documents govern most of them. RFC 9700 gives current best practice for OAuth 2.0 security, and the MCP specification requires a protected MCP server to act as an OAuth 2.1 resource server. For cloud platforms, Microsoft documents Conditional Access as its policy engine that combines signals into if-then access decisions. Authorization details change over time, so cite the version of the specification you used. The MCP specification page read for this guide is the 2025-06-18 revision, and it requires protected MCP servers to publish protected resource metadata. Sources: [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700); [MCP specification: authorization](https://modelcontextprotocol.io/specification/2025-06-18/basic/authorization); [Microsoft Entra Conditional Access overview](https://learn.microsoft.com/en-us/entra/identity/conditional-access/overview). ## How do you use the mapping in practice? Build a table with one row per framework item and one column per layer, and mark the control you have in each cell. Empty cells are your gaps. Ask vendors to mark the same table from their documents. Compare the answers with what the documentation states, not what a sales deck says. Review the table when a source updates. NIST says the AI RMF 1.0 is being revised, and OWASP and MITRE both update their lists, so set a quarterly reminder to check the sources and note the date read. Sources: [Zero Trust overview, Microsoft Learn](https://learn.microsoft.com/en-us/security/zero-trust/zero-trust-overview). ## Sources - [OWASP Top 10 for Agentic Applications for 2026](https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/) (genai.owasp.org), read 2026-11-16 - [OWASP Top 10 for LLM and Gen AI Apps (2025)](https://genai.owasp.org/llm-top-10/) (genai.owasp.org), read 2026-11-16 - [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) (nist.gov), read 2026-11-16 - [NIST AI 600-1, the GenAI Profile](https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf) (nvlpubs.nist.gov), read 2026-11-16 - [MITRE ATLAS](https://atlas.mitre.org/) (atlas.mitre.org), read 2026-11-16 - [MCP specification: security best practices](https://modelcontextprotocol.io/specification/2025-06-18/basic/security_best_practices) (modelcontextprotocol.io), read 2026-11-16 - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems) (nist.gov), read 2026-11-16 - [NIST AI 100-2 E2025, Adversarial Machine Learning](https://csrc.nist.gov/pubs/ai/100/2/e2025/final) (csrc.nist.gov), read 2026-11-16 - [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700) (rfc-editor.org), read 2026-11-16 - [MCP specification: authorization](https://modelcontextprotocol.io/specification/2025-06-18/basic/authorization) (modelcontextprotocol.io), read 2026-11-16 - [Microsoft Entra Conditional Access overview](https://learn.microsoft.com/en-us/entra/identity/conditional-access/overview) (learn.microsoft.com), read 2026-11-16 - [Zero Trust overview, Microsoft Learn](https://learn.microsoft.com/en-us/security/zero-trust/zero-trust-overview) (learn.microsoft.com), read 2026-11-16 ## Related pages - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [OAuth scope (glossary)](/glossary/oauth-scope/) - [All guides](/guides/) - [AI Agent Security for Small Teams: Which Layer First](/guides/ai-agent-security-for-small-teams-which-layer-first/) - [AI Agent Security Incident Response Playbook](/guides/ai-agent-security-incident-response-playbook/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security incident response playbook URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-incident-response-playbook/ Updated: 2026-11-16 When an AI agent misbehaves, contain it first, revoke its credentials second, preserve its trail third, then find the cause and restore with tighter limits. Decide in advance who can pull the kill switch, which logs hold the agent's tool calls and how you cut tokens at the identity provider. ## What do you do when an agent misbehaves? Stop the agent from acting, then work out what it did. Pause the session or cut its tool access before any analysis, because a running agent keeps acting. Straiker says its kill switch cuts off tool access, suspends the session, isolates or takes an agent offline. CrowdStrike says its AI Kill Switch directs connected AI gateways. Check what your product can pause, and rehearse it. Decide in advance who may pull the switch and under what conditions. A written rule such as any agent that reads credentials outside its scope is paused at once spares a debate at 2 a.m. Record the contact for each platform owner. Sources: [Straiker: agentic kill switch](https://www.straiker.ai/solution/agentic-kill-switch); [CrowdStrike: AI Kill Switch](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/). ## How do you revoke an agent and review its trail? Revoke at the identity provider and at each tool, then export the trail. Revoking only the agent leaves tokens that other services still honor. Island describes Okta and Entra revocation steps for agent identities, and Zenity says it reads agent identities from Entra ID and Okta. Microsoft documents that Conditional Access for agents evaluates agent, user, device and risk signals to allow, block or require controls. Revoke in the right order: stop the agent, revoke its tokens, rotate any secret it could read, then disable the agent identity. Check that services which cached the token also stop honoring it, since some keep a token valid until it expires. Sources: [Island: agentic identity](https://www.island.io/product/ai/agentic-identity); [Zenity: agentic identity and access management](https://zenity.io/platform/agentic-identity-and-access-management); [Microsoft Entra: Conditional Access for agents](https://learn.microsoft.com/en-us/entra/identity/conditional-access/agent-id). ## Where is the evidence? The evidence sits in the agent platform's logs, the endpoint product's decision log, the gateway and the identity provider's sign-in logs. List them before an incident. Bay describes session timelines for investigation, and Cisco says every block, approval and decision is recorded with the agent, sandbox and policy behind it. Export the records before retention rules remove them. Preserve evidence before you change settings. Snapshot the agent configuration, the list of connected tools and the recent logs. If the platform keeps logs for a short time, export first and investigate second. Sources: [Bay: AI agent investigation](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide); [Cisco: protect enterprise coding agents with DefenseClaw](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell). ## How do you find the cause? Trace the chain from the input to the action: what the agent read, what it decided, which tool it called and with what permission. The cause is usually one of OWASP's three: too much functionality, too much permission or too much autonomy. Triggers include indirect prompt injection and, in multi-agent systems, a compromised peer agent. Ask whether a person, a document, a web page or another agent supplied the input that changed the behavior. Where the input is external, add it to your blocked sources and test that the new rule fires. Sources: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/); [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems). ## How do you restore safely? Restore with narrower tools, narrower scopes and an approval step for the action that went wrong. Test the new limit with a harmless case. Do not restore the old credential. Issue a new short-lived one. Tell the users and owners of the agent what happened and what changed. If data left the company, follow your normal breach process; an agent incident is still a data incident. Sources: [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700). ## What do you write down afterward? Record the timeline, the control that should have stopped it, the control that did, the gap and the owner of the fix. Add the incident to the monthly report. Use MITRE ATLAS technique names where they fit, so the record uses shared words. Rehearse once a year with a tabletop. Use a real agent in your estate, a plausible failure and the actual contacts. Fix the gaps the exercise finds in the playbook itself. Sources: [MITRE ATLAS](https://atlas.mitre.org/). ## Sources - [Straiker: agentic kill switch](https://www.straiker.ai/solution/agentic-kill-switch) (straiker.ai), read 2026-11-16 - [CrowdStrike: AI Kill Switch](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/) (crowdstrike.com), read 2026-11-16 - [Island: agentic identity](https://www.island.io/product/ai/agentic-identity) (island.io), read 2026-11-16 - [Zenity: agentic identity and access management](https://zenity.io/platform/agentic-identity-and-access-management) (zenity.io), read 2026-11-16 - [Microsoft Entra: Conditional Access for agents](https://learn.microsoft.com/en-us/entra/identity/conditional-access/agent-id) (learn.microsoft.com), read 2026-11-16 - [Bay: AI agent investigation](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide) (bay.io), read 2026-11-16 - [Cisco: protect enterprise coding agents with DefenseClaw](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell) (blogs.cisco.com), read 2026-11-16 - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems) (nist.gov), read 2026-11-16 - [RFC 9700, Best Current Practice for OAuth 2.0 Security](https://www.rfc-editor.org/rfc/rfc9700) (rfc-editor.org), read 2026-11-16 - [MITRE ATLAS](https://atlas.mitre.org/) (atlas.mitre.org), read 2026-11-16 ## Related pages - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [AI Agent Security for Multi-Agent Systems](/guides/ai-agent-security-for-multi-agent-systems/) - [Kill switch (glossary)](/glossary/kill-switch/) - [All guides](/guides/) - [AI Agent Security Frameworks Mapped to Five Layers](/guides/ai-agent-security-frameworks-mapped-to-layers/) - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security layers explained URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-layers-explained/ Updated: 2026-11-16 AI agent security splits into five layers: endpoint, browser, cloud workspace, identity and gateway. Each sits in a different place, sees different events and blocks different actions. Cover first the layer where your agents already run, then add the layer that closes the biggest gap, using vendor documents read on 16 Nov 2026. ## What are the five layers of AI agent security? The five layers are the endpoint, the browser, the cloud workspace, identity and the gateway. They are five places where a control can see an agent act, and no one place sees them all. OWASP's Excessive Agency entry names three root causes that every layer tries to limit: excessive functionality, excessive permissions and excessive autonomy. Controls at each layer cut one or more of the three. Treat the layers as a coverage grid rather than a ladder. A control at one layer does not make the next one unnecessary, and a gap at one layer can be closed only at another. A coding agent on a laptop, a browser agent filling a web form and a Copilot Studio agent in a business unit are three different agents with three different control points. Sources: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/). ## What does the endpoint layer see? The endpoint layer sees agents running on laptops and developer machines: desktop assistants, coding agents, command-line agents and local MCP servers. A control here can read the tool call before the agent executes it. Claude Code's hooks show the mechanism: a PreToolUse hook runs before a tool call executes and can block it. Vendors build on such hooks. Pillar Security says its Runtime Guardian sits in the pre-tool hook of Claude Code, Codex CLI and Cursor, and Operant says its Endpoint Protector inspects AI IDEs, coding agents, desktop and MCP clients. Local controls have one more advantage: they work when the laptop is off the corporate network. Pillar Security says its endpoint decisions run locally and keep working offline. Ask every endpoint vendor where the decision is made, on the device or in a cloud service, and what happens when the connection drops. Sources: [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks); [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint); [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector). ## What does the browser layer see? The browser layer sees AI websites, AI extensions and agents that act inside a browser, such as filling forms or downloading files. Network inspection alone sees the traffic but not the page actions. Vendors place controls in an extension or a managed browser. Cato says its client covers browser AI and desktop AI apps and can monitor, anonymize or block a prompt, and Island builds agent controls around its enterprise browser, including a headless browser for agents. Browser controls also cover the human side of AI use. Staff paste text into chat sites, upload files and install AI extensions, and those are the actions a browser control can see and stop. List the AI extensions installed in your fleet before you evaluate any product. Sources: [Cato Networks: AI Endpoint](https://www.catonetworks.com/platform/cato-ai-endpoint/); [Island: agentic identity](https://www.island.io/product/ai/agentic-identity). ## What do the cloud workspace layer and identity layer add? The cloud workspace layer covers agents that makers build inside SaaS and cloud platforms, and the identity layer covers the credentials and permissions every agent carries. Neither runs on a device you manage. For Copilot Studio, Microsoft documents that agents can be represented as Entra identities and governed with Conditional Access once Agent 365 is onboarded. AWS documents a guardrail setting you attach to a Bedrock agent. Both show that the platform is a control point of its own. Platform controls differ by platform, so a team using three agent platforms may manage three policy surfaces. A third-party product can unify them with one inventory and one policy language, but check whether it only discovers agents on each platform or can block actions there. Sources: [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance); [Amazon Bedrock: associate a guardrail with your agent](https://docs.aws.amazon.com/bedrock/latest/userguide/agents-guardrail.html). ## What does the gateway layer add? A gateway adds one control point on the path between agents and models, tools or MCP servers. It sees every call routed through it and sees nothing that bypasses it. Palo Alto Networks documents an AI Gateway that sits between users or applications and each LLM provider. Cisco publishes a plugin that inspects LLM and MCP requests on a Kong gateway. A gateway fits server-side agents well and fits laptops less well. Fail-closed behavior matters at a gateway. If the gateway is unreachable, a fail-closed plugin blocks the call and a fail-open plugin lets it through. Cisco documents fail-closed defaults for its checks; confirm the same for any gateway you adopt. Sources: [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview); [Cisco AI Defense plugin for Kong Gateway](https://github.com/cisco-ai-defense/ai-defense-kong-plugin). ## Which layer should come first? Start with the layer where your agents run today and where a block is cheapest to deploy. For many companies that is the endpoint, because developers already run coding agents and an MDM script is the install path. If your agents live in Copilot Studio or Bedrock, start with the cloud workspace and identity layers instead. Add the gateway when agents call external models and tools from servers. Confirm the picks against the Layer Coverage Map before you buy. Keep a one-page layer map for your own estate: agents by layer, the control at each, the owner and the date last tested. Update it when a new agent platform is approved. The Layer Coverage Map on this site shows what each vendor documents so you can fill the map without starting from blank. Sources: [What is Amazon Bedrock AgentCore](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html); [Bay: agent runtime enforcement](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide). ## Sources - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks) (docs.anthropic.com), read 2026-11-16 - [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint) (pillar.security), read 2026-11-16 - [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector) (operant.ai), read 2026-11-16 - [Cato Networks: AI Endpoint](https://www.catonetworks.com/platform/cato-ai-endpoint/) (catonetworks.com), read 2026-11-16 - [Island: agentic identity](https://www.island.io/product/ai/agentic-identity) (island.io), read 2026-11-16 - [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance) (learn.microsoft.com), read 2026-11-16 - [Amazon Bedrock: associate a guardrail with your agent](https://docs.aws.amazon.com/bedrock/latest/userguide/agents-guardrail.html) (docs.aws.amazon.com), read 2026-11-16 - [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview) (docs.paloaltonetworks.com), read 2026-11-16 - [Cisco AI Defense plugin for Kong Gateway](https://github.com/cisco-ai-defense/ai-defense-kong-plugin) (github.com), read 2026-11-16 - [What is Amazon Bedrock AgentCore](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/what-is-bedrock-agentcore.html) (docs.aws.amazon.com), read 2026-11-16 - [Bay: agent runtime enforcement](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide) (bay.io), read 2026-11-16 ## Related pages - [AI Agent Security Platform or Point Solutions?](/guides/ai-agent-security-platform-or-point-solutions/) - [AI Agent Security for Small Teams: Which Layer First](/guides/ai-agent-security-for-small-teams-which-layer-first/) - [AI Agent Security Roles: Who Owns Each Layer](/guides/ai-agent-security-roles-who-owns-each-layer/) - [Agent runtime (glossary)](/glossary/agent-runtime/) - [AI browser (glossary)](/glossary/ai-browser/) - [All guides](/guides/) - [AI Agent Security Incident Response Playbook](/guides/ai-agent-security-incident-response-playbook/) - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security metrics and reporting URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-metrics-and-reporting/ Updated: 2026-11-16 Track AI agent security with a small set of counts: agents found, share covered by a blocking control, actions blocked versus only alerted, credential age, time to revoke and unapproved tools. A monthly report lists each against last month, names open gaps by layer and carries the date each control was last tested. ## What should you measure? Measure coverage, enforcement, credentials and response time. Coverage is the share of known agents under a control, enforcement is blocks against alerts, credentials is how old and how broad they are, and response is minutes to revoke. NIST's AI RMF asks organizations to measure and manage AI risk, which gives these counts a governance home. Set a target for each measure and an owner. A coverage target of 95 percent of managed endpoints with the hook installed is concrete, and a note that says improve coverage is not. Targets can start modest and tighten every quarter. - Agents inventoried, and agents with an owner. - Share of managed endpoints with the enforcement hook installed. - Actions blocked, asked, and only alerted, by layer. - Age and scope of agent credentials; share that are short-lived. - Minutes from decision to revoked access. - Unapproved MCP servers and tools found and closed. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## Where do the numbers come from? They come from the vendor consoles and the logs you forward. Check which export format each product offers before you plan a dashboard. Bay says events reach the customer SIEM in OCSF format, Operant says Endpoint Protector forwards telemetry to Splunk, Datadog, Elastic and Chronicle, and Cisco says decisions can go to Splunk or any OpenTelemetry backend. OpenTelemetry's GenAI conventions name agent spans, which helps you join data across tools. If a vendor offers no export, ask for a report on a schedule or an API. Several vendors list SIEM or OpenTelemetry export in their documentation, and others list none found, which is a factor when you pick between products with similar blocking. Sources: [Bay: AI agent investigation](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide); [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector); [Cisco: protect enterprise coding agents with DefenseClaw](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell); [OpenTelemetry GenAI semantic conventions](https://opentelemetry.io/docs/specs/semconv/gen-ai/). ## What belongs on a monthly report? A monthly report fits on one page: five counts against last month, the top three open gaps by layer, incidents with time to contain, and the date each control was last tested. Add one line on policy changes and one on exceptions granted. Reviewers want to know what changed, not the whole inventory. Keep a standing section on exceptions. Count the exceptions granted, who approved them and when each expires. An exception with no expiry is a permanent gap that nobody remembers. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## How do you count agents you did not know about? Count discovered agents separately from approved ones, and track the gap. A rising discovered count with a flat approved count means shadow agents are growing. Microsoft's release notes describe local AI agent discovery on Windows and macOS that reports each agent's configured MCP servers, which gives an endpoint-side count. Set a rule for what happens to a newly found agent: assign an owner within a week, approve or block within two. Report how many sit past that deadline. Sources: [What's new in Microsoft Defender for Endpoint](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint). ## Which metrics mislead? Alert counts mislead because a noisy rule inflates them and a silent gap hides. Pair every alert count with the block count and the coverage share. Vendor dashboards also count only what the product covers. Compare against your own list of agents by layer. Add a measure for the time between a vendor's feature release and your use of it. Agent security products change monthly, and unused controls count as no coverage. Sources: [Zero Trust overview, Microsoft Learn](https://learn.microsoft.com/en-us/security/zero-trust/zero-trust-overview). ## How often should you review the targets? Review them quarterly, and after any new agent platform or a major vendor change. Keep the definitions stable so the trend means something. Record each definition change with its date on the report. Share a short version with leadership: coverage, blocks, incidents and the largest open gap, in plain words. Keep the detail for the security team. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## Sources - [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) (nist.gov), read 2026-11-16 - [Bay: AI agent investigation](https://www.bay.io/ai-agent-investigation?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide) (bay.io), read 2026-11-16 - [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector) (operant.ai), read 2026-11-16 - [Cisco: protect enterprise coding agents with DefenseClaw](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell) (blogs.cisco.com), read 2026-11-16 - [OpenTelemetry GenAI semantic conventions](https://opentelemetry.io/docs/specs/semconv/gen-ai/) (opentelemetry.io), read 2026-11-16 - [What's new in Microsoft Defender for Endpoint](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint) (learn.microsoft.com), read 2026-11-16 - [Zero Trust overview, Microsoft Learn](https://learn.microsoft.com/en-us/security/zero-trust/zero-trust-overview) (learn.microsoft.com), read 2026-11-16 ## Related pages - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [AI Agent Security Incident Response Playbook](/guides/ai-agent-security-incident-response-playbook/) - [AI Agent Security Roles: Who Owns Each Layer](/guides/ai-agent-security-roles-who-owns-each-layer/) - [Agent observability (glossary)](/glossary/agent-observability/) - [All guides](/guides/) - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Platform or Point Solutions?](/guides/ai-agent-security-platform-or-point-solutions/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security platform or point solutions URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-platform-or-point-solutions/ Updated: 2026-11-16 An AI agent security platform covers several layers under one policy and one inventory, while a point solution covers one layer well. One product is enough when your agents run mostly in one or two places; spread across endpoints, SaaS and servers, expect to combine tools or accept gaps the documentation shows. ## What is an AI agent security platform? An AI agent security platform is one product that discovers agents and enforces policy across more than one layer. The shared parts are an inventory, a policy engine and an audit trail. Zenity says one runtime engine evaluates every agent action and can let it through, modify it, block it or shut the agent down. Straiker says its Discover AI finds sanctioned and shadow AI across endpoint, browser, gateway, cloud and connected systems. Counting layers is a quick screen, not a verdict. A platform that lists five layers may mean five integrations of different depth. Read the documentation for the verb at each layer before you count it, and check whether the claim is confirmed by a source beyond the vendor. Sources: [Zenity: runtime boundaries](https://zenity.io/platform/runtime-boundaries); [Straiker: Discover AI](https://www.straiker.ai/products/discover-ai). ## What is the difference between a platform and a point solution? A platform trades depth for reach, and a point solution trades reach for depth. A point solution such as a network control point can be strong on one path and silent on the rest. WitnessAI enforces at the network and says it needs no endpoint client; Pillar Security concentrates on pre-tool hooks for coding agents on managed machines. Each documents one strong control point, not all five layers. A point solution can also be the right second purchase. If a platform covers your endpoints well and you later add a gateway for server-side agents, a gateway specialist or a gateway you already run may fit better than extending the first vendor. Sources: [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/); [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint). ## When does one product cover enough layers? One product is enough when your agents run in one or two of the five layers and the product documents both. A company whose agents are all Copilot Studio and Microsoft 365 agents may need the Microsoft stack and little else. Microsoft documents a registry, Entra identities with Conditional Access, and runtime protection in one control plane for such agents. The documents also state license needs, so price the combination. The other direction also holds. A team with agents on laptops, in the browser and in two SaaS platforms is unlikely to find one product that documents strong blocking on all four. Plan for two products and one shared inventory, or for one product and a written list of accepted gaps. Sources: [Microsoft Agent 365 overview](https://learn.microsoft.com/en-us/microsoft-agent-365/overview); [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance). ## What gaps should you expect in a platform? Expect uneven depth. A platform may discover agents on a layer and block on another. Operant says its Microsoft Copilot path observes, and other paths block inline. Read each vendor's documentation for the verbs: discovers, observes, alerts, blocks. Then check which layers have the strongest verbs for your first deployment. Ask for the integration list in writing: your identity provider, your EDR, your SIEM and your ticketing tool. Several vendors publish integrations for some of these and none found for others, so a platform with no integration for your SIEM costs integration work that no price sheet shows. Sources: [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector); [Operant AI: integrations](https://www.operant.ai/platform/integrations). ## How do platform owners differ from startups? Platform owners such as Microsoft, CrowdStrike, Cisco and Palo Alto Networks bundle agent security into stacks you may already license. Startups ship faster on one new surface. CrowdStrike says its AI gateway is included with Guardian at no additional cost. Microsoft lists Agent 365 at $15.00 per user per month, billed yearly. Most startups publish no price. Procurement differs too. Platform owners usually sell through enterprise agreements and publish list prices for some products. Startups usually quote after a demo. Ask early for a pilot with a stated end date and success measures, since few publish a price you can model. Sources: [CrowdStrike: AI Kill Switch](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/); [Microsoft Agent 365](https://www.microsoft.com/en-us/microsoft-agent-365). ## How do you decide? List your agents by layer, mark the top two layers, then shortlist vendors that document strong blocking on both. If no one does, buy the strongest on the first layer and plan the second. Ask each vendor for the deployment path in writing, the supported operating systems and the time to first inventory. Document the decision and the reason, so the next review can see what changed. If a vendor adds a layer later, reopen the question with the new documentation. Sources: [Check Point (Lakera) docs: agent discovery](https://docs.lakera.ai/docs/agent-security/discovery). ## Sources - [Zenity: runtime boundaries](https://zenity.io/platform/runtime-boundaries) (zenity.io), read 2026-11-16 - [Straiker: Discover AI](https://www.straiker.ai/products/discover-ai) (straiker.ai), read 2026-11-16 - [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/) (witness.ai), read 2026-11-16 - [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint) (pillar.security), read 2026-11-16 - [Microsoft Agent 365 overview](https://learn.microsoft.com/en-us/microsoft-agent-365/overview) (learn.microsoft.com), read 2026-11-16 - [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance) (learn.microsoft.com), read 2026-11-16 - [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector) (operant.ai), read 2026-11-16 - [Operant AI: integrations](https://www.operant.ai/platform/integrations) (operant.ai), read 2026-11-16 - [CrowdStrike: AI Kill Switch](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/) (crowdstrike.com), read 2026-11-16 - [Microsoft Agent 365](https://www.microsoft.com/en-us/microsoft-agent-365) (microsoft.com), read 2026-11-16 - [Check Point (Lakera) docs: agent discovery](https://docs.lakera.ai/docs/agent-security/discovery) (docs.lakera.ai), read 2026-11-16 ## Related pages - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Solutions Market Map, November 2026](/guides/ai-agent-security-solutions-market-map/) - [AI Agent Security Roles: Who Owns Each Layer](/guides/ai-agent-security-roles-who-owns-each-layer/) - [Agent registry (glossary)](/glossary/agent-registry/) - [All guides](/guides/) - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security risks, by layer URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-risks/ Updated: 2026-11-16 The main AI agent security risks are indirect prompt injection, excessive agency, over-broad credentials, risky tools and MCP servers, sensitive data leaving, and runaway consumption. Each sits at a layer where a control is best placed: tool hooks at the endpoint, scopes in identity, allow lists at the gateway and platform settings in the cloud workspace. ## What are the main AI agent security risks? The main risks are manipulated inputs, excessive agency, misused credentials, an untrusted tool supply chain, data leaving, and unbounded consumption. NIST's request for information on securing AI agent systems groups them as models meeting adversarial data such as indirect prompt injection, use of insecure models, and models taking harmful actions without any attack. OWASP's 2025 LLM list names prompt injection, exposure of sensitive information, supply chain, excessive agency and unbounded consumption, and its Top 10 for Agentic Applications for 2026 was built with more than 100 experts for agents that plan and act. Not every risk needs a new product. Some are old problems with new reach: an over-privileged service account is dangerous whether a person or an agent holds it. What changes is speed and scale, because an agent can try many actions in minutes and chain tools a person would not combine. Sources: [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems); [OWASP Top 10 for LLM and Gen AI Apps (2025)](https://genai.owasp.org/llm-top-10/); [OWASP Top 10 for Agentic Applications for 2026](https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/). ## Which risks sit at which layer? Each risk is best stopped where the harmful step becomes visible. A credential read is visible on the endpoint, a form upload in the browser, an over-broad token in identity, a tool call at the gateway and a connector setting in the cloud workspace. The list below pairs each risk with its natural layer. Many risks need two layers, such as a gateway block backed by a scoped token. Pick the three risks that match your agents and write them as one-line scenarios with an owner. A coding-agent team might choose credential theft from the workstation, a poisoned MCP server and an unreviewed package install. A SaaS-builder team might choose an over-shared connector, a runaway loop and data sent to a public model. - Indirect prompt injection: gateway or endpoint inspection of tool inputs and outputs. - Excessive agency: identity scopes and endpoint approval steps. - Credential and secret theft: endpoint hooks and short-lived credentials. - Risky MCP servers and tools: gateway allow list and endpoint inventory. - Data leaving through uploads and pastes: browser controls. - Unbounded consumption: gateway limits and platform quotas. Sources: [OWASP Top 10 for LLM and Gen AI Apps (2025)](https://genai.owasp.org/llm-top-10/); [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/). ## How does excessive agency turn into harm? Excessive agency turns a bad output into a real action. OWASP says it enables damaging actions in response to unexpected, ambiguous or manipulated model outputs, whatever caused the malfunction. Triggers include hallucination, direct or indirect prompt injection, a malicious extension and, in multi-agent systems, a compromised peer agent. The fix is fewer tools, narrower permissions and approval for high-impact actions. Hallucination belongs on the list because it needs no attacker. An agent that invents a file path or a command argument can delete the wrong thing. OWASP notes that excessive agency can follow from a poorly performing model as well as from an attack, so limits apply even when you trust every input. Sources: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/). ## What are the risks from tools and MCP servers? Tools and MCP servers extend what an agent can do, so a compromised or over-privileged one extends the damage. The MCP specification documents a confused deputy problem in MCP proxy servers that connect to third-party APIs. Treat each server as a dependency: approve it, authorize it, scope it and log its calls. Approving a server is not a one-time event. Track the publisher, the version, the permissions it asks for and the data it can return. A server that was safe at approval can change with an update, so re-check on every version bump. Sources: [MCP specification: security best practices](https://modelcontextprotocol.io/specification/2025-06-18/basic/security_best_practices); [MCP specification: authorization](https://modelcontextprotocol.io/specification/2025-06-18/basic/authorization). ## Where can you read attacker techniques for AI systems? MITRE ATLAS is a public knowledge base of adversary tactics and techniques involving AI. On 16 Nov 2026 its home page counted 16 tactics, 221 techniques, 44 mitigations and 77 case studies. NIST's adversarial machine learning taxonomy, AI 100-2 E2025, adds a shared vocabulary for attack types and mitigations. Use both to name risks in your own register. Use technique names in tickets and detections so that analysts, vendors and auditors share terms. When a vendor says it detects a technique, ask for the technique identifier and the layer where it sees the behavior. Sources: [MITRE ATLAS](https://atlas.mitre.org/); [NIST AI 100-2 E2025, Adversarial Machine Learning](https://csrc.nist.gov/pubs/ai/100/2/e2025/final). ## What do the risks mean for buying? Write each risk as a question for the vendor: at which layer does the control act, and does it stop the action or only alert? Vendors differ widely on both. The Agent Control Point Map shows the point in an agent's run where each vendor acts, from before start to after the fact. Ask also what the product records when it cannot decide. A control that quietly allows an unknown action leaves a blind spot, while one that asks or denies by default leaves a work-around that people will want to remove. Pick the default on purpose for each action class. Sources: [Zenity: runtime boundaries](https://zenity.io/platform/runtime-boundaries). ## Sources - [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems) (nist.gov), read 2026-11-16 - [OWASP Top 10 for LLM and Gen AI Apps (2025)](https://genai.owasp.org/llm-top-10/) (genai.owasp.org), read 2026-11-16 - [OWASP Top 10 for Agentic Applications for 2026](https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/) (genai.owasp.org), read 2026-11-16 - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [MCP specification: security best practices](https://modelcontextprotocol.io/specification/2025-06-18/basic/security_best_practices) (modelcontextprotocol.io), read 2026-11-16 - [MCP specification: authorization](https://modelcontextprotocol.io/specification/2025-06-18/basic/authorization) (modelcontextprotocol.io), read 2026-11-16 - [MITRE ATLAS](https://atlas.mitre.org/) (atlas.mitre.org), read 2026-11-16 - [NIST AI 100-2 E2025, Adversarial Machine Learning](https://csrc.nist.gov/pubs/ai/100/2/e2025/final) (csrc.nist.gov), read 2026-11-16 - [Zenity: runtime boundaries](https://zenity.io/platform/runtime-boundaries) (zenity.io), read 2026-11-16 ## Related pages - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [AI Agent Security Testing and Red Team Exercises](/guides/ai-agent-security-testing-and-red-team-exercises/) - [Goal hijacking (glossary)](/glossary/goal-hijacking/) - [Memory poisoning (glossary)](/glossary/memory-poisoning/) - [Unbounded consumption (glossary)](/glossary/unbounded-consumption/) - [All guides](/guides/) - [AI Agent Security Platform or Point Solutions?](/guides/ai-agent-security-platform-or-point-solutions/) - [AI Agent Security Roles: Who Owns Each Layer](/guides/ai-agent-security-roles-who-owns-each-layer/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security roles: who owns each layer URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-roles-who-owns-each-layer/ Updated: 2026-11-16 Each layer of AI agent security has a natural owner: endpoint and browser sit with IT or endpoint security, identity with the identity team, the gateway with platform or network engineering, and the cloud workspace with the business platform admins. Security sets policy across all five and reviews results. ## Who owns the endpoint layer, identity layer and gateway layer? Endpoint security or IT owns the endpoint layer, the identity team owns the identity layer, and platform or network engineering owns the gateway layer. Security owns the policy that all three enforce. The owner is whoever already runs the install path. Vendors publish MDM and EDR paths for endpoint tools, so the team that runs the MDM runs the rollout. Write the owner next to each layer in the layer map. If two teams each think the other owns the browser extension, nobody patches it. A named owner and a named backup close that gap. Sources: [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint); [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector). ## Who owns the browser and cloud workspace layers? Endpoint or IT teams own the browser layer, as it ships as an extension or managed browser. Platform admins own the cloud workspace layer. For Copilot Studio, Microsoft documents that admins use data policies in the Power Platform admin center to govern agent capabilities, so those admins hold the controls. Business-unit makers build many cloud workspace agents. They need training and clear limits, since they are not security staff. Admins can use data policies to limit connectors and channels before a maker builds. Sources: [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance). ## How do security, IT and engineering share the work? Security writes the policy and reviews, IT and identity teams deploy, and engineering owns the agents it builds. A shared register names one owner per agent and per control. NIST's AI RMF puts governance first, with named roles and accountability, which this split follows. Hold a short monthly meeting between the layer owners and the policy owner. The agenda is the monthly report, the open gaps and the decisions needed. Keep it to thirty minutes. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## Who owns each agent? Every agent needs a named human owner, because incident response starts there. Check Point says its discovery records each agent's owner, model, tools and MCP servers. Microsoft's Agent 365 registry lists agents with owner, usage and risk. Orphaned agents should be disabled. When an owner leaves, reassign their agents before the account is disabled. An agent that loses its owner keeps running with its credentials, which is how orphaned agents appear. Sources: [Check Point: AI agent security](https://www.checkpoint.com/ai-security/ai-agent-security/); [Microsoft Agent 365 overview](https://learn.microsoft.com/en-us/microsoft-agent-365/overview). ## What should the policy owner decide? The policy owner decides which agents and tools are approved, which actions need a human, and who may grant exceptions. Keep these few and written. Conditional Access in Entra shows the pattern: policies are if-then rules that combine signals and are owned by one team. Record who may approve each class of agent and tool, and keep the approvals in a log that auditors can read. Approvals tied to named roles survive staff turnover. Sources: [Microsoft Entra Conditional Access overview](https://learn.microsoft.com/en-us/entra/identity/conditional-access/overview). ## How do you avoid gaps between owners? Gaps appear at the edges: a browser extension that reaches a SaaS agent, or a token issued by one team and used by another. Review those edges quarterly. Use one inventory across the layers, and make each control owner sign off on the layer map. Include the vendor relationship in the ownership table: who holds the contract, who gets the security notices and who reviews the documentation each quarter. Sources: [Zero Trust overview, Microsoft Learn](https://learn.microsoft.com/en-us/security/zero-trust/zero-trust-overview). ## Sources - [Pillar Security: Agentic Endpoint](https://www.pillar.security/platform/agentic-endpoint) (pillar.security), read 2026-11-16 - [Operant AI: Endpoint Protector](https://www.operant.ai/platform/endpoint-protector) (operant.ai), read 2026-11-16 - [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance) (learn.microsoft.com), read 2026-11-16 - [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) (nist.gov), read 2026-11-16 - [Check Point: AI agent security](https://www.checkpoint.com/ai-security/ai-agent-security/) (checkpoint.com), read 2026-11-16 - [Microsoft Agent 365 overview](https://learn.microsoft.com/en-us/microsoft-agent-365/overview) (learn.microsoft.com), read 2026-11-16 - [Microsoft Entra Conditional Access overview](https://learn.microsoft.com/en-us/entra/identity/conditional-access/overview) (learn.microsoft.com), read 2026-11-16 - [Zero Trust overview, Microsoft Learn](https://learn.microsoft.com/en-us/security/zero-trust/zero-trust-overview) (learn.microsoft.com), read 2026-11-16 ## Related pages - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [AI Agent Security for Small Teams: Which Layer First](/guides/ai-agent-security-for-small-teams-which-layer-first/) - [Agent lifecycle (glossary)](/glossary/agent-lifecycle/) - [All guides](/guides/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [AI Agent Security Solutions Market Map, November 2026](/guides/ai-agent-security-solutions-market-map/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security solutions market map URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-solutions-market-map/ Updated: 2026-11-16 Vendors selling AI agent security in November 2026 fall into five groups: endpoint and EDR vendors, network and SASE-linked vendors, an enterprise browser maker, agent-specialist startups, and cloud platform owners. Each group starts at a different layer, and the vendor pages here show how far each has extended. ## Which kinds of vendors sell agent security? Five kinds sell it, and each started at a different layer. Endpoint vendors started on the device, network vendors on the traffic path, the browser maker in the browser, specialists on the agent, and cloud owners on their own platform. A group is a starting point, not a limit: several vendors now document three or four layers. The map is a guide to origins, which explain strengths and gaps. A vendor from the endpoint side tends to document the device in detail, and a vendor from the network side tends to document the traffic path. Expect each to be thinner where it started elsewhere. Sources: [SentinelOne: Prompt Security agentic AI security](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/); [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview). ## How do endpoint vendors approach it? Endpoint vendors use their sensors and hooks. CrowdStrike says agent-level hooks for Claude Code, OpenAI Codex, GitHub Copilot, OpenCode and Cursor block before harmful actions run, and SentinelOne describes an AI endpoint agent and browser extension through Prompt Security. Microsoft's Defender release notes describe local AI agent discovery on Windows and macOS. The Defender release notes read for this guide describe local AI agent discovery as generally available on Windows and macOS. That is discovery; confirm in the documentation which blocking actions are generally available and which are still preview before you plan around them. Sources: [CrowdStrike: Falcon Guardian](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/); [SentinelOne: Prompt Security agentic AI security](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/); [What's new in Microsoft Defender for Endpoint](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint). ## How do network and SASE-linked vendors approach it? They add agent controls to products that already sit on the traffic path. Cato says its AI Endpoint client finds local agents, and Palo Alto Networks documents an AI Gateway between applications and each LLM provider. Cisco publishes a plugin for Kong Gateway that inspects LLM and MCP requests, and WitnessAI enforces an approved tool list at the network. Network-linked vendors reach agents that run on servers and in cloud accounts, where no endpoint sensor sits. They reach laptops less directly unless they ship a client, which Cato does through its AI Endpoint product. Sources: [Cato Networks: AI Endpoint](https://www.catonetworks.com/platform/cato-ai-endpoint/); [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview); [Cisco AI Defense plugin for Kong Gateway](https://github.com/cisco-ai-defense/ai-defense-kong-plugin); [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/). ## How do agent-specialist startups approach it? Specialists start with the agent itself and then widen. Zenity and Noma Security name SaaS and cloud agent platforms they cover, Lasso Security discovers agents agentlessly, Straiker documents all five layers, and Bay decides on the device with no proxy. Most publish no price and keep documentation behind a login, so the evidence is vendor pages and third-party reports. Several specialists have been bought or are being bought by larger vendors, so check the ownership line on a vendor page before you sign a multi-year contract. The profiles here state the owner as the vendor's pages and third-party reports give it on the read date. Sources: [Zenity: runtime boundaries](https://zenity.io/platform/runtime-boundaries); [Noma Security: SaaS agents](https://noma.security/platform/saas-agents); [Lasso Security: AI gateway security](https://www.lasso.security/use-cases/ai-gateway-security); [Bay: agent runtime enforcement](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide). ## How do platform owners differ from startups? Platform owners control the agent platform, so their controls can sit inside it. Microsoft documents Agent 365, Entra identities and Purview audit for Copilot Studio agents, and AWS documents guardrails you attach to a Bedrock agent. Startups add cross-platform inventory and one policy across platforms. Compare what the platform gives free with what a vendor adds. A third group matters to the map: tools that were built for application or LLM security and now extend to agents. Their strength is prompt and output inspection, and their agent-action control varies, so read the verbs. Sources: [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance); [Amazon Bedrock: associate a guardrail with your agent](https://docs.aws.amazon.com/bedrock/latest/userguide/agents-guardrail.html); [Microsoft Agent 365 overview](https://learn.microsoft.com/en-us/microsoft-agent-365/overview). ## What changed in 2026? Vendors added agent-specific products and endpoint hooks during 2026. CrowdStrike announced an agentic identity provider on 2 Sep 2026 and an AI Kill Switch on 3 Nov 2026, and Operant announced Browser Protection on 27 Aug 2026. Expect more releases; each profile on this site shows its read date. Out of scope here are agent identity platforms, governance platforms and gateway products ranked on their own. This site scores how agent security solutions cover those layers, not the stand-alone categories. Sources: [CrowdStrike: Agentic Identity Provider](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/); [CrowdStrike: AI Kill Switch](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/); [Operant AI: Semantic Firewall and Browser Protection](https://www.operant.ai/art-kubed/introducing-operant-semantic-firewall). ## Sources - [SentinelOne: Prompt Security agentic AI security](https://www.sentinelone.com/platform/prompt-security-agentic-ai-security/) (sentinelone.com), read 2026-11-16 - [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview) (docs.paloaltonetworks.com), read 2026-11-16 - [CrowdStrike: Falcon Guardian](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/) (crowdstrike.com), read 2026-11-16 - [What's new in Microsoft Defender for Endpoint](https://learn.microsoft.com/en-us/defender-endpoint/whats-new-in-microsoft-defender-endpoint) (learn.microsoft.com), read 2026-11-16 - [Cato Networks: AI Endpoint](https://www.catonetworks.com/platform/cato-ai-endpoint/) (catonetworks.com), read 2026-11-16 - [Cisco AI Defense plugin for Kong Gateway](https://github.com/cisco-ai-defense/ai-defense-kong-plugin) (github.com), read 2026-11-16 - [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/) (witness.ai), read 2026-11-16 - [Zenity: runtime boundaries](https://zenity.io/platform/runtime-boundaries) (zenity.io), read 2026-11-16 - [Noma Security: SaaS agents](https://noma.security/platform/saas-agents) (noma.security), read 2026-11-16 - [Lasso Security: AI gateway security](https://www.lasso.security/use-cases/ai-gateway-security) (lasso.security), read 2026-11-16 - [Bay: agent runtime enforcement](https://www.bay.io/agent-runtime-enforcement?utm_source=aiagentsecuritysolutions.com&utm_medium=referral&utm_campaign=guide) (bay.io), read 2026-11-16 - [Microsoft Copilot Studio security and governance](https://learn.microsoft.com/en-us/microsoft-copilot-studio/security-and-governance) (learn.microsoft.com), read 2026-11-16 - [Amazon Bedrock: associate a guardrail with your agent](https://docs.aws.amazon.com/bedrock/latest/userguide/agents-guardrail.html) (docs.aws.amazon.com), read 2026-11-16 - [Microsoft Agent 365 overview](https://learn.microsoft.com/en-us/microsoft-agent-365/overview) (learn.microsoft.com), read 2026-11-16 - [CrowdStrike: Agentic Identity Provider](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/) (crowdstrike.com), read 2026-11-16 - [CrowdStrike: AI Kill Switch](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/) (crowdstrike.com), read 2026-11-16 - [Operant AI: Semantic Firewall and Browser Protection](https://www.operant.ai/art-kubed/introducing-operant-semantic-firewall) (operant.ai), read 2026-11-16 ## Related pages - [AI Agent Security Platform or Point Solutions?](/guides/ai-agent-security-platform-or-point-solutions/) - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [AI Agent Security Frameworks Mapped to Five Layers](/guides/ai-agent-security-frameworks-mapped-to-layers/) - [Agent sprawl (glossary)](/glossary/agent-sprawl/) - [All guides](/guides/) - [AI Agent Security Roles: Who Owns Each Layer](/guides/ai-agent-security-roles-who-owns-each-layer/) - [AI Agent Security Testing and Red Team Exercises](/guides/ai-agent-security-testing-and-red-team-exercises/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security testing and red team exercises URL: https://aiagentsecuritysolutions.com/guides/ai-agent-security-testing-and-red-team-exercises/ Updated: 2026-11-16 Test an agent security control by giving the agent a harmless task that the control should stop, in a test environment, and confirming the block, the log and the alert. Run exercises per layer, use public technique catalogs to choose cases, and record every gap with an owner. ## How do you test an agent security control? Pick one control, define the action it should stop, run a harmless version of that action in a test setup and check three things: it was blocked, it was logged and someone was told. If any of the three is missing, the control is not working. Use test accounts, test data and a sandbox. Never test with production credentials. Write the expected result before the test. A block test has three pass conditions, and a test with no stated expectation proves nothing. Keep each case short enough that another engineer can repeat it. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## Which attacks should an exercise include? Include the risk classes NIST and OWASP name: manipulated inputs, excessive agency, credential misuse, risky tools and data leaving. Write each as a scenario, not a script. NIST's request for information groups risks as adversarial data, insecure models and harmful actions without attack. MITRE ATLAS lists 221 techniques and 77 case studies to pick from. Prefer realistic scenarios over exotic ones. Most losses start with an over-privileged token or an unreviewed tool, so test those first, then widen to manipulated inputs. Record the layer where each scenario should be stopped. Sources: [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems); [MITRE ATLAS](https://atlas.mitre.org/); [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/). ## What do you test at each layer? Test the point where the layer acts. On the endpoint, test that a pre-tool check blocks an out-of-scope file read. In the browser, test that an upload to an unapproved AI site is stopped. In the cloud workspace, test the platform guardrail. In identity, test that a revoked token fails. At the gateway, test that an unlisted tool is denied. Test the failure modes too. Unplug the network from a test laptop and see whether the endpoint control still decides. Make the gateway unreachable and see whether calls fail closed or open. Sources: [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks); [Amazon Bedrock: associate a guardrail with your agent](https://docs.aws.amazon.com/bedrock/latest/userguide/agents-guardrail.html); [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/). ## What tools exist for red teaming agents? A few are public. Operant publishes Woodpecker, an open-source red teaming tool, on GitHub under Apache-2.0. Check the license and run any tool only in a test environment. Vendor products often include their own test modes; ask for the documentation. When you run a vendor's own test mode, ask which actions it covers and which it leaves out. A vendor demo shows the cases the product handles, and your suite should include the ones it does not. Sources: [Operant AI Woodpecker on GitHub](https://github.com/OperantAI/woodpecker). ## How do you handle findings? Log each finding with the layer, the control, the gap, the severity and an owner. Fix the control, not the test. Retest after the fix and add the case to a standing suite that runs after every policy change. Share results with the vendor if a control fails. Vendors fix what they hear about, and a documented gap helps your next renewal conversation. Sources: [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework). ## How often should you run exercises? Run a small suite monthly and a full exercise twice a year, or after any new agent platform. Agents and their tools change often, so a control that passed last quarter can fail today. Record each run in the monthly report. Include a purple-team step: the people who run the test sit with the people who run detection and walk through what each saw. Gaps in what the logs show are often the most useful finding. Sources: [MITRE ATLAS](https://atlas.mitre.org/). ## Sources - [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) (nist.gov), read 2026-11-16 - [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems) (nist.gov), read 2026-11-16 - [MITRE ATLAS](https://atlas.mitre.org/) (atlas.mitre.org), read 2026-11-16 - [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/) (genai.owasp.org), read 2026-11-16 - [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks) (docs.anthropic.com), read 2026-11-16 - [Amazon Bedrock: associate a guardrail with your agent](https://docs.aws.amazon.com/bedrock/latest/userguide/agents-guardrail.html) (docs.aws.amazon.com), read 2026-11-16 - [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/) (witness.ai), read 2026-11-16 - [Operant AI Woodpecker on GitHub](https://github.com/OperantAI/woodpecker) (github.com), read 2026-11-16 ## Related pages - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [Guardrail policy (glossary)](/glossary/guardrail-policy/) - [All guides](/guides/) - [AI Agent Security Solutions Market Map, November 2026](/guides/ai-agent-security-solutions-market-map/) - [The ranking](/ranking/) - [Glossary](/glossary/) --- # AI agent security glossary URL: https://aiagentsecuritysolutions.com/glossary/ Updated: 2026-11-16 Twenty-eight terms define the vocabulary of AI agent security, from agent runtime, agent registry and non-human identity to kill switch, blast radius, memory poisoning and goal hijacking. Each entry gives a short definition and one sourced example read on 16 Nov 2026. ## 28 terms - [A2A protocol](/glossary/a2a-protocol/): The A2A protocol, short for Agent2Agent, is an open specification that lets agents built on different frameworks discover each other's capabilities and exchange tasks and messages. - [Access request workflow](/glossary/access-request-workflow/): An access request workflow is the path by which a user or agent asks for something policy blocks, and an administrator approves or denies it. - [Agent credential vault](/glossary/agent-credential-vault/): An agent credential vault is a managed store for the keys, tokens and secrets an agent needs, kept outside the agent so it never holds them in plain view. - [Agent framework](/glossary/agent-framework/): An agent framework is a code library that supplies the building blocks for an agent: model calls, tool definitions, memory and control flow. - [Agent lifecycle](/glossary/agent-lifecycle/): The agent lifecycle is the sequence from creating an agent through approval, change, monitoring and retirement. - [Agent observability](/glossary/agent-observability/): Agent observability is the collection of prompts, responses, tool calls and outcomes that show what an agent did and why. - [Agent orchestration](/glossary/agent-orchestration/): Agent orchestration is the logic that decides which agent or tool acts next, passes state between steps and handles retries. - [Agent registry](/glossary/agent-registry/): An agent registry is a central list of the agents, tools and skills in use, with an owner and an approval status for each. - [Agent runtime](/glossary/agent-runtime/): An agent runtime is the software environment in which an AI agent executes its loop of model calls, tool calls and state handling. - [Agent sprawl](/glossary/agent-sprawl/): Agent sprawl is the spread of AI agents across platforms, SDKs and employee machines without a central record of who owns them or what they can reach. - [Agent trace](/glossary/agent-trace/): An agent trace is the ordered record of one agent session: the prompts, the tool calls, the results and the downstream actions they caused. - [AI browser](/glossary/ai-browser/): An AI browser is a web browser with a built-in AI assistant that can read pages, fill forms and take actions for the user. - [Autonomy level](/glossary/autonomy-level/): An autonomy level states how much an agent may do on its own, from suggesting actions to acting without any approval. - [Blast radius](/glossary/blast-radius/): Blast radius is the amount of data, systems and other agents that one compromised or misbehaving agent can reach. - [Goal hijacking](/glossary/goal-hijacking/): Goal hijacking is an attack in which hostile content changes what an agent is trying to do, so it pursues the attacker's aim instead of the assigned task. - [Guardrail policy](/glossary/guardrail-policy/): A guardrail policy is a written rule that maps a detected condition to an action, such as alert, block, redact or log. - [Just-in-time access](/glossary/just-in-time-access/): Just-in-time access grants an identity a permission only when a task needs it and removes it when the task ends. - [Kill switch](/glossary/kill-switch/): A kill switch is a control that stops an agent, cuts its access or isolates it when risk is high. - [LLM gateway](/glossary/llm-gateway/): An LLM gateway is a proxy that sits between applications or agents and the AI models they call. - [Memory poisoning](/glossary/memory-poisoning/): Memory poisoning is the planting of false or hostile instructions in the long-term memory an agent keeps between sessions. - [Model routing](/glossary/model-routing/): Model routing is the practice of sending each request to a model chosen by cost, task or policy. - [Multi-agent system](/glossary/multi-agent-system/): A multi-agent system is a set of AI agents that divide a task and pass work, context and results to each other. - [Non-human identity](/glossary/non-human-identity/): A non-human identity is an account, key, token or certificate used by software or a machine instead of a person. - [OAuth scope](/glossary/oauth-scope/): An OAuth scope is a label in an OAuth request that names the kind of access a token grants, such as read-only access to one resource. - [Permission boundary](/glossary/permission-boundary/): A permission boundary is a ceiling on the permissions an identity can receive, set apart from the policies that grant them. - [SaaS agent](/glossary/saas-agent/): A SaaS agent is an AI agent that runs inside a software-as-a-service platform, using that platform's data, tools and identity model. - [Unbounded consumption](/glossary/unbounded-consumption/): Unbounded consumption is runaway use of model calls, tokens or compute, either by accident or on purpose. - [Workspace agent](/glossary/workspace-agent/): A workspace agent is an AI assistant or agent embedded in productivity and collaboration software, with delegated access to a user's documents, mail and chats. ## Related pages - [Guides](/guides/) - [FAQ](/faq/) - [The ranking](/ranking/) - [Layer Coverage Map](/layer-coverage-map/) - [Agent Control Point Map](/agent-control-point-map/) --- # A2A protocol: definition and example URL: https://aiagentsecuritysolutions.com/glossary/a2a-protocol/ Updated: 2026-11-16 The A2A protocol, short for Agent2Agent, is an open specification that lets agents built on different frameworks discover each other's capabilities and exchange tasks and messages. It is separate from the protocols that connect an agent to tools. For security, it adds agent-to-agent traffic that a gateway or registry has to recognize. ## Example The A2A Protocol site announced that A2A joins the Agentic AI Foundation, a governance move recorded on the specification site on 16 Nov 2026. ## Sources - [A2A Protocol specification](https://a2a-protocol.org/latest/specification/) (a2a-protocol.org), read 2026-11-16 ## Related pages - [Multi-agent system](/glossary/multi-agent-system/) - [Agent registry](/glossary/agent-registry/) - [LLM gateway](/glossary/llm-gateway/) - [AI Agent Security for Multi-Agent Systems](/guides/ai-agent-security-for-multi-agent-systems/) - [All glossary terms](/glossary/) - [Workspace agent](/glossary/workspace-agent/) - [Access request workflow](/glossary/access-request-workflow/) - [Guides](/guides/) --- # Access request workflow: definition and example URL: https://aiagentsecuritysolutions.com/glossary/access-request-workflow/ Updated: 2026-11-16 An access request workflow is the path by which a user or agent asks for something policy blocks, and an administrator approves or denies it. It keeps controls usable without opening everything by default. The decision and its reason should be logged. ## Example Harmonic's documentation says a user's attempt to use an MCP server without permission is blocked and creates an access request for an administrator. ## Sources - [Configure MCP server access controls](https://docs.harmonicsecurity.app/portal-guides/managing-your-organization/configure-mcp-server-access-controls) (docs.harmonicsecurity.app), read 2026-11-16 ## Related pages - [Agent registry](/glossary/agent-registry/) - [Autonomy level](/glossary/autonomy-level/) - [Just-in-time access](/glossary/just-in-time-access/) - [All glossary terms](/glossary/) - [A2A protocol](/glossary/a2a-protocol/) - [Agent credential vault](/glossary/agent-credential-vault/) - [Guides](/guides/) --- # Agent credential vault: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-credential-vault/ Updated: 2026-11-16 An agent credential vault is a managed store for the keys, tokens and secrets an agent needs, kept outside the agent so it never holds them in plain view. The agent asks for access and the vault releases a credential for a stated purpose. A vault only helps if the agent's own role cannot read all of it. ## Example Zenity says its AgentCorruption research retrieved API keys, OAuth tokens and other credentials stored in AWS Secrets Manager and in environment variables from AgentCore agents. ## Sources - [Zenity Labs reports AgentCorruption](https://zenity.io/press-release/zenity-labs-discloses-agentcorruption-a-chain-of-aws-agentcore-flaws) (zenity.io), read 2026-11-16 ## Related pages - [Non-human identity](/glossary/non-human-identity/) - [Just-in-time access](/glossary/just-in-time-access/) - [Blast radius](/glossary/blast-radius/) - [All glossary terms](/glossary/) - [Access request workflow](/glossary/access-request-workflow/) - [Agent framework](/glossary/agent-framework/) - [Guides](/guides/) --- # Agent framework: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-framework/ Updated: 2026-11-16 An agent framework is a code library that supplies the building blocks for an agent: model calls, tool definitions, memory and control flow. Teams adopt one so they do not rebuild the loop for each agent. For security, framework use is a clue to finding agents that never appear in a vendor catalog. ## Example CrowdStrike says planned Falcon Guardian coverage of cloud workloads and Kubernetes extends to enterprise-developed agents, including those built with LangChain and LangGraph, and describes that coverage as pre-beta. ## Sources - [CrowdStrike advances AI security and safety with Falcon Guardian](https://www.crowdstrike.com/en-us/blog/crowdstrike-advances-ai-security-and-safety-falcon-guardian/) (crowdstrike.com), read 2026-11-16 ## Related pages - [Agent runtime](/glossary/agent-runtime/) - [Agent orchestration](/glossary/agent-orchestration/) - [Agent registry](/glossary/agent-registry/) - [All glossary terms](/glossary/) - [Agent credential vault](/glossary/agent-credential-vault/) - [Agent lifecycle](/glossary/agent-lifecycle/) - [Guides](/guides/) --- # Agent lifecycle: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-lifecycle/ Updated: 2026-11-16 The agent lifecycle is the sequence from creating an agent through approval, change, monitoring and retirement. Each stage needs an owner and a record. Retirement matters most, because forgotten agents keep their access. ## Example Microsoft's Entra Agent ID documentation says agents receive the same identity-driven protections as users and workloads, including adaptive access policies, real-time risk detection and lifecycle management. ## Sources - [What is Microsoft Entra Agent ID?](https://learn.microsoft.com/en-us/entra/agent-id/what-is-microsoft-entra-agent-id) (learn.microsoft.com), read 2026-11-16 ## Related pages - [Agent registry](/glossary/agent-registry/) - [Non-human identity](/glossary/non-human-identity/) - [Agent sprawl](/glossary/agent-sprawl/) - [AI Agent Security Roles: Who Owns Each Layer](/guides/ai-agent-security-roles-who-owns-each-layer/) - [All glossary terms](/glossary/) - [Agent framework](/glossary/agent-framework/) - [Agent observability](/glossary/agent-observability/) - [Guides](/guides/) --- # Agent observability: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-observability/ Updated: 2026-11-16 Agent observability is the collection of prompts, responses, tool calls and outcomes that show what an agent did and why. It supports audit, investigation and cost control. It records activity; it does not by itself stop an action. ## Example Operant says Microsoft Purview already records Copilot interactions for audit and compliance, and that its integration turns that record into security telemetry attributed to the user, app surface and resources referenced. ## Sources - [Operant: coverage for Microsoft Copilot and embedded AI assistants](https://www.operant.ai/art-kubed/announcing-operants-coverage-for-microsoft-copilot-and-every-ai-assistant-embedded-in-microsoft-365) (operant.ai), read 2026-11-16 ## Related pages - [Agent trace](/glossary/agent-trace/) - [Workspace agent](/glossary/workspace-agent/) - [Guardrail policy](/glossary/guardrail-policy/) - [AI Agent Security Metrics and Monthly Reporting](/guides/ai-agent-security-metrics-and-reporting/) - [All glossary terms](/glossary/) - [Agent lifecycle](/glossary/agent-lifecycle/) - [Agent orchestration](/glossary/agent-orchestration/) - [Guides](/guides/) --- # Agent orchestration: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-orchestration/ Updated: 2026-11-16 Agent orchestration is the logic that decides which agent or tool acts next, passes state between steps and handles retries. In some designs the orchestration is explicit code; in others a model picks the next step. Whoever sets the orchestration also sets where a person or a policy check can interrupt the work. ## Example LangChain describes its LangGraph product with the phrase balance agent control with agency, which is the orchestration trade-off in one line. ## Sources - [LangGraph by LangChain](https://www.langchain.com/langgraph) (langchain.com), read 2026-11-16 ## Related pages - [Multi-agent system](/glossary/multi-agent-system/) - [Agent framework](/glossary/agent-framework/) - [Autonomy level](/glossary/autonomy-level/) - [All glossary terms](/glossary/) - [Agent observability](/glossary/agent-observability/) - [Agent registry](/glossary/agent-registry/) - [Guides](/guides/) --- # Agent registry: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-registry/ Updated: 2026-11-16 An agent registry is a central list of the agents, tools and skills in use, with an owner and an approval status for each. Discovery feeds it and policy reads from it. Without one, approval is a document nobody can enforce. ## Example Noma says every agent, MCP server and skill it finds on an endpoint receives a status of approved, needs review or blocked, and that newly discovered assets enter the registry automatically. ## Sources - [Noma Extends Agent Security and Governance to the Employee Endpoint](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html) (prnewswire.com), read 2026-11-16 ## Related pages - [Agent sprawl](/glossary/agent-sprawl/) - [Agent lifecycle](/glossary/agent-lifecycle/) - [Access request workflow](/glossary/access-request-workflow/) - [AI Agent Security Platform or Point Solutions?](/guides/ai-agent-security-platform-or-point-solutions/) - [All glossary terms](/glossary/) - [Agent orchestration](/glossary/agent-orchestration/) - [Agent runtime](/glossary/agent-runtime/) - [Guides](/guides/) --- # Agent runtime: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-runtime/ Updated: 2026-11-16 An agent runtime is the software environment in which an AI agent executes its loop of model calls, tool calls and state handling. It decides which files, network destinations, processes and credentials the agent can reach. A control inside the runtime can see and stop individual actions, while a control outside it mostly sees traffic. ## Example Cisco describes NVIDIA OpenShell as the secure runtime that governs how agents execute, which resources and data they can reach, and where inference is routed. ## Sources - [Protect enterprise coding agents with Cisco DefenseClaw and NVIDIA OpenShell](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell) (blogs.cisco.com), read 2026-11-16 ## Related pages - [Agent framework](/glossary/agent-framework/) - [Blast radius](/glossary/blast-radius/) - [Kill switch](/glossary/kill-switch/) - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [All glossary terms](/glossary/) - [Agent registry](/glossary/agent-registry/) - [Agent sprawl](/glossary/agent-sprawl/) - [Guides](/guides/) --- # Agent sprawl: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-sprawl/ Updated: 2026-11-16 Agent sprawl is the spread of AI agents across platforms, SDKs and employee machines without a central record of who owns them or what they can reach. It grows because agents can be built in many places with little review. The first control against it is a single registry fed by discovery from several layers. ## Example Microsoft says agents can be created through Microsoft platforms, integrated through partner platforms and SDKs, or introduced locally without a central deployment process. ## Sources - [Agents are your newest insiders. Security must move at their speed.](https://www.microsoft.com/en-us/security/blog/2026/10/19/agents-are-your-newest-insiders-security-must-move-at-their-speed/) (microsoft.com), read 2026-11-16 ## Related pages - [Agent registry](/glossary/agent-registry/) - [Agent lifecycle](/glossary/agent-lifecycle/) - [Non-human identity](/glossary/non-human-identity/) - [AI Agent Security Solutions Market Map, November 2026](/guides/ai-agent-security-solutions-market-map/) - [All glossary terms](/glossary/) - [Agent runtime](/glossary/agent-runtime/) - [Agent trace](/glossary/agent-trace/) - [Guides](/guides/) --- # Agent trace: definition and example URL: https://aiagentsecuritysolutions.com/glossary/agent-trace/ Updated: 2026-11-16 An agent trace is the ordered record of one agent session: the prompts, the tool calls, the results and the downstream actions they caused. Investigators use it to see which input led to which action. A good trace links steps across tools, not just model calls. ## Example CrowdStrike says Falcon Guardian reconstructs agent sessions and downstream execution into a causal investigation. ## Sources - [Falcon Guardian defines the next generation of AI security](https://www.crowdstrike.com/en-us/blog/falcon-guardian-defines-next-generation-of-ai-security/) (crowdstrike.com), read 2026-11-16 ## Related pages - [Agent observability](/glossary/agent-observability/) - [Kill switch](/glossary/kill-switch/) - [Goal hijacking](/glossary/goal-hijacking/) - [All glossary terms](/glossary/) - [Agent sprawl](/glossary/agent-sprawl/) - [AI browser](/glossary/ai-browser/) - [Guides](/guides/) --- # AI browser: definition and example URL: https://aiagentsecuritysolutions.com/glossary/ai-browser/ Updated: 2026-11-16 An AI browser is a web browser with a built-in AI assistant that can read pages, fill forms and take actions for the user. It differs from an AI extension because the agent is part of the browser itself. Security controls have to recognize it by name or by behavior. ## Example Island says its extension supports AI browsers including Atlas and Comet, with the same feature support it has in Chrome and Edge. ## Sources - [Island: enterprise browser for agents](https://www.island.io/product/ai/enterprise-browser-for-agents) (island.io), read 2026-11-16 ## Related pages - [SaaS agent](/glossary/saas-agent/) - [Workspace agent](/glossary/workspace-agent/) - [Agent sprawl](/glossary/agent-sprawl/) - [AI Agent Security Layers Explained: Five Places to Cover](/guides/ai-agent-security-layers-explained/) - [All glossary terms](/glossary/) - [Agent trace](/glossary/agent-trace/) - [Autonomy level](/glossary/autonomy-level/) - [Guides](/guides/) --- # Autonomy level: definition and example URL: https://aiagentsecuritysolutions.com/glossary/autonomy-level/ Updated: 2026-11-16 An autonomy level states how much an agent may do on its own, from suggesting actions to acting without any approval. Teams set it per agent or per action type. Higher levels need stronger logging and faster blocking. ## Example Noma says agents can chain actions together across a session without human approval at every step. ## Sources - [Noma Extends Agent Security and Governance to the Employee Endpoint](https://www.prnewswire.com/news-releases/noma-extends-agent-security-and-governance-to-the-employee-endpoint-302890925.html) (prnewswire.com), read 2026-11-16 ## Related pages - [Kill switch](/glossary/kill-switch/) - [Access request workflow](/glossary/access-request-workflow/) - [Agent orchestration](/glossary/agent-orchestration/) - [All glossary terms](/glossary/) - [AI browser](/glossary/ai-browser/) - [Blast radius](/glossary/blast-radius/) - [Guides](/guides/) --- # Blast radius: definition and example URL: https://aiagentsecuritysolutions.com/glossary/blast-radius/ Updated: 2026-11-16 Blast radius is the amount of data, systems and other agents that one compromised or misbehaving agent can reach. It is set by the agent's credentials, network access and shared infrastructure. Smaller roles and separate identities shrink it. ## Example Zenity says an AgentCore default role extended to all agents in an AWS account and region, so one public-facing agent opened a path to the rest. ## Sources - [Zenity Labs reports AgentCorruption](https://zenity.io/press-release/zenity-labs-discloses-agentcorruption-a-chain-of-aws-agentcore-flaws) (zenity.io), read 2026-11-16 ## Related pages - [Permission boundary](/glossary/permission-boundary/) - [Kill switch](/glossary/kill-switch/) - [Memory poisoning](/glossary/memory-poisoning/) - [AI Agent Security for Small Teams: Which Layer First](/guides/ai-agent-security-for-small-teams-which-layer-first/) - [All glossary terms](/glossary/) - [Autonomy level](/glossary/autonomy-level/) - [Goal hijacking](/glossary/goal-hijacking/) - [Guides](/guides/) --- # Goal hijacking: definition and example URL: https://aiagentsecuritysolutions.com/glossary/goal-hijacking/ Updated: 2026-11-16 Goal hijacking is an attack in which hostile content changes what an agent is trying to do, so it pursues the attacker's aim instead of the assigned task. It often arrives through files and web content the agent reads. Defenses check the agent's actions against the task, not only its inputs. ## Example Prompt Security, now part of SentinelOne, published research on 20 August 2026 covering an AGENTS.md goal hijack and arguing that approval prompts should show the real destination and payload. ## Sources - [When AI agents become the supply chain](https://prompt.security/blog/when-ai-agents-become-the-supply-chain-hidden-control-planes-in-agentic-systems) (prompt.security), read 2026-11-16 ## Related pages - [Memory poisoning](/glossary/memory-poisoning/) - [Autonomy level](/glossary/autonomy-level/) - [SaaS agent](/glossary/saas-agent/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [All glossary terms](/glossary/) - [Blast radius](/glossary/blast-radius/) - [Guardrail policy](/glossary/guardrail-policy/) - [Guides](/guides/) --- # Guardrail policy: definition and example URL: https://aiagentsecuritysolutions.com/glossary/guardrail-policy/ Updated: 2026-11-16 A guardrail policy is a written rule that maps a detected condition to an action, such as alert, block, redact or log. It is separate from the detector that finds the condition. Teams tune the mapping per agent and per data class. ## Example Cisco says its Project Quarry detection language is in production in AI Defense and maps findings to alert, block, redact, quarantine or log. ## Sources - [Project Quarry: agent security detections](https://blogs.cisco.com/ai/project-quarry-agent-security-detections) (blogs.cisco.com), read 2026-11-16 ## Related pages - [Agent observability](/glossary/agent-observability/) - [LLM gateway](/glossary/llm-gateway/) - [Autonomy level](/glossary/autonomy-level/) - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [AI Agent Security Testing and Red Team Exercises](/guides/ai-agent-security-testing-and-red-team-exercises/) - [All glossary terms](/glossary/) - [Goal hijacking](/glossary/goal-hijacking/) - [Just-in-time access](/glossary/just-in-time-access/) - [Guides](/guides/) --- # Just-in-time access: definition and example URL: https://aiagentsecuritysolutions.com/glossary/just-in-time-access/ Updated: 2026-11-16 Just-in-time access grants an identity a permission only when a task needs it and removes it when the task ends. For an AI agent this replaces standing access that a user once approved. It limits what a hijacked agent can do after the task window closes. ## Example Microsoft says its Dynamic Permissions let an organization scope an agent's authority to the task, make it time-bound, and contract or remove it as the task changes or completes. ## Sources - [Agents are your newest insiders. Security must move at their speed.](https://www.microsoft.com/en-us/security/blog/2026/10/19/agents-are-your-newest-insiders-security-must-move-at-their-speed/) (microsoft.com), read 2026-11-16 ## Related pages - [OAuth scope](/glossary/oauth-scope/) - [Permission boundary](/glossary/permission-boundary/) - [Access request workflow](/glossary/access-request-workflow/) - [All glossary terms](/glossary/) - [Guardrail policy](/glossary/guardrail-policy/) - [Kill switch](/glossary/kill-switch/) - [Guides](/guides/) --- # Kill switch: definition and example URL: https://aiagentsecuritysolutions.com/glossary/kill-switch/ Updated: 2026-11-16 A kill switch is a control that stops an agent, cuts its access or isolates it when risk is high. It differs from blocking one action because it ends the agent's whole run. It only works if it sits where the agent cannot disable it. ## Example SiliconANGLE reported that Microsoft's Conditional Access for agents can hold higher-impact actions for human approval as risk rises and, at the top of the scale, isolate or stop the agent. ## Sources - [Microsoft targets the agentic insider with new controls for rogue AI agents](https://siliconangle.com/2026/10/19/microsoft-targets-the-agentic-insider-with-new-controls-for-rogue-ai-agents/) (siliconangle.com), read 2026-11-16 ## Related pages - [Blast radius](/glossary/blast-radius/) - [Agent trace](/glossary/agent-trace/) - [Autonomy level](/glossary/autonomy-level/) - [AI Agent Security Incident Response Playbook](/guides/ai-agent-security-incident-response-playbook/) - [All glossary terms](/glossary/) - [Just-in-time access](/glossary/just-in-time-access/) - [LLM gateway](/glossary/llm-gateway/) - [Guides](/guides/) --- # LLM gateway: definition and example URL: https://aiagentsecuritysolutions.com/glossary/llm-gateway/ Updated: 2026-11-16 An LLM gateway is a proxy that sits between applications or agents and the AI models they call. It can inspect prompts and responses, apply policy, meter usage and route requests. It sees model traffic, not what an agent does with a response afterward. ## Example CrowdStrike says developers connect an existing SDK or coding tool to Falcon AI Gateway by changing the base URL to point at the gateway and using a gateway key. ## Sources - [Inside Falcon AI Gateway: One Control Point for AI Traffic](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/) (crowdstrike.com), read 2026-11-16 ## Related pages - [Model routing](/glossary/model-routing/) - [Guardrail policy](/glossary/guardrail-policy/) - [A2A protocol](/glossary/a2a-protocol/) - [All glossary terms](/glossary/) - [Kill switch](/glossary/kill-switch/) - [Memory poisoning](/glossary/memory-poisoning/) - [Guides](/guides/) --- # Memory poisoning: definition and example URL: https://aiagentsecuritysolutions.com/glossary/memory-poisoning/ Updated: 2026-11-16 Memory poisoning is the planting of false or hostile instructions in the long-term memory an agent keeps between sessions. Because memory is read again later, one successful write can steer many future conversations. Defenses limit who can write memory and watch what it contains. ## Example Zenity says researchers created malicious memories in AWS AgentCore that directed an agent's future conversations to an attacker-controlled destination. ## Sources - [Zenity Labs reports AgentCorruption](https://zenity.io/press-release/zenity-labs-discloses-agentcorruption-a-chain-of-aws-agentcore-flaws) (zenity.io), read 2026-11-16 ## Related pages - [Goal hijacking](/glossary/goal-hijacking/) - [Blast radius](/glossary/blast-radius/) - [Agent trace](/glossary/agent-trace/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [All glossary terms](/glossary/) - [LLM gateway](/glossary/llm-gateway/) - [Model routing](/glossary/model-routing/) - [Guides](/guides/) --- # Model routing: definition and example URL: https://aiagentsecuritysolutions.com/glossary/model-routing/ Updated: 2026-11-16 Model routing is the practice of sending each request to a model chosen by cost, task or policy. It lets teams keep sensitive prompts on approved models and cheaper tasks on cheaper ones. The router is also a control point that can log and enforce. ## Example WitnessAI announced on 15 September 2026 AI FinOps capabilities that combine multi-provider metering, model routing, shadow AI discovery and runtime security. ## Sources - [WitnessAI Introduces AI FinOps Capabilities](https://www.prnewswire.com/news-releases/witnessai-introduces-ai-finops-capabilities-to-control-enterprise-ai-spend-and-drive-effective-roi-302878268.html) (prnewswire.com), read 2026-11-16 ## Related pages - [LLM gateway](/glossary/llm-gateway/) - [Unbounded consumption](/glossary/unbounded-consumption/) - [Guardrail policy](/glossary/guardrail-policy/) - [All glossary terms](/glossary/) - [Memory poisoning](/glossary/memory-poisoning/) - [Multi-agent system](/glossary/multi-agent-system/) - [Guides](/guides/) --- # Multi-agent system: definition and example URL: https://aiagentsecuritysolutions.com/glossary/multi-agent-system/ Updated: 2026-11-16 A multi-agent system is a set of AI agents that divide a task and pass work, context and results to each other. Each handoff is a place where identity, scope and data can be checked, and a place where untrusted content can travel from one agent to the next. Securing it means controlling the links as well as the agents. ## Example OWASP's Securing Agentic Applications V2 guide says it covers secure architectures, identity and access controls, runtime protection and assurance testing across single-agent and multi-agent environments. ## Sources - [Securing Agentic Applications V2](https://genai.owasp.org/resource/securing-agentic-applications-v2/) (genai.owasp.org), read 2026-11-16 ## Related pages - [A2A protocol](/glossary/a2a-protocol/) - [Agent orchestration](/glossary/agent-orchestration/) - [Blast radius](/glossary/blast-radius/) - [AI Agent Security for Multi-Agent Systems](/guides/ai-agent-security-for-multi-agent-systems/) - [All glossary terms](/glossary/) - [Model routing](/glossary/model-routing/) - [Non-human identity](/glossary/non-human-identity/) - [Guides](/guides/) --- # Non-human identity: definition and example URL: https://aiagentsecuritysolutions.com/glossary/non-human-identity/ Updated: 2026-11-16 A non-human identity is an account, key, token or certificate used by software or a machine instead of a person. AI agents add a new group because they hold credentials, call tools and act for a user. Managing them means inventory, ownership, short lifetimes and revocation. ## Example The OWASP Non-Human Identity Top 10 for 2025 lists long-lived secrets as NHI7, one of ten risks it names for application developers. ## Sources - [OWASP Non-Human Identity Top 10 2025](https://owasp.org/www-project-non-human-identities-top-10/2025/) (owasp.org), read 2026-11-16 ## Related pages - [Agent credential vault](/glossary/agent-credential-vault/) - [Just-in-time access](/glossary/just-in-time-access/) - [Agent lifecycle](/glossary/agent-lifecycle/) - [All glossary terms](/glossary/) - [Multi-agent system](/glossary/multi-agent-system/) - [OAuth scope](/glossary/oauth-scope/) - [Guides](/guides/) --- # OAuth scope: definition and example URL: https://aiagentsecuritysolutions.com/glossary/oauth-scope/ Updated: 2026-11-16 An OAuth scope is a label in an OAuth request that names the kind of access a token grants, such as read-only access to one resource. A narrow scope limits what a connected agent can do even if its token leaks. Broad scopes turn one approval into standing access to a whole account. ## Example RFC 6749 defines the OAuth 2.0 framework as a way for a third-party application to obtain limited access to an HTTP service, and it has a section on access token scope. ## Sources - [RFC 6749: The OAuth 2.0 Authorization Framework](https://www.rfc-editor.org/rfc/rfc6749) (rfc-editor.org), read 2026-11-16 ## Related pages - [Permission boundary](/glossary/permission-boundary/) - [Just-in-time access](/glossary/just-in-time-access/) - [Non-human identity](/glossary/non-human-identity/) - [AI Agent Security Frameworks Mapped to Five Layers](/guides/ai-agent-security-frameworks-mapped-to-layers/) - [All glossary terms](/glossary/) - [Guides](/guides/) --- # Permission boundary: definition and example URL: https://aiagentsecuritysolutions.com/glossary/permission-boundary/ Updated: 2026-11-16 A permission boundary is a ceiling on the permissions an identity can receive, set apart from the policies that grant them. Even if someone attaches a broad policy, the effective permissions cannot exceed the boundary. For agents it keeps one over-broad role from reaching every other agent. ## Example AWS documentation states that a permissions boundary uses a managed policy to set the maximum permissions an identity-based policy can grant to an IAM entity. ## Sources - [Permissions boundaries for IAM entities](https://docs.aws.amazon.com/IAM/latest/UserGuide/access_policies_boundaries.html) (docs.aws.amazon.com), read 2026-11-16 ## Related pages - [OAuth scope](/glossary/oauth-scope/) - [Blast radius](/glossary/blast-radius/) - [Just-in-time access](/glossary/just-in-time-access/) - [AI Agent Security Best Practices, Layer by Layer](/guides/ai-agent-security-best-practices/) - [All glossary terms](/glossary/) - [SaaS agent](/glossary/saas-agent/) - [Guides](/guides/) --- # SaaS agent: definition and example URL: https://aiagentsecuritysolutions.com/glossary/saas-agent/ Updated: 2026-11-16 A SaaS agent is an AI agent that runs inside a software-as-a-service platform, using that platform's data, tools and identity model. Because the agent runs in the vendor's cloud, no process exists on an employee machine for an endpoint tool to watch. Controls must come from the platform, an API integration or a gateway. ## Example Zenity Labs reported that its SalesBleed research found Salesforce Agentforce flaws in which a poisoned lead could steer an agent to leak CRM data without a click. ## Sources - [SalesBleed: 0-click data exfiltration on Agentforce](https://labs.zenity.io/post/salesbleed-0-click-data-exfiltration-on-agentforce) (labs.zenity.io), read 2026-11-16 ## Related pages - [Workspace agent](/glossary/workspace-agent/) - [Goal hijacking](/glossary/goal-hijacking/) - [Blast radius](/glossary/blast-radius/) - [All glossary terms](/glossary/) - [Permission boundary](/glossary/permission-boundary/) - [Unbounded consumption](/glossary/unbounded-consumption/) - [Guides](/guides/) --- # Unbounded consumption: definition and example URL: https://aiagentsecuritysolutions.com/glossary/unbounded-consumption/ Updated: 2026-11-16 Unbounded consumption is runaway use of model calls, tokens or compute, either by accident or on purpose. An agent that repeats work or picks the costliest model for every step can run up cost at machine speed. Limits and budgets per agent keep it visible. ## Example Operant describes an agent that re-reads a full repository when it needed one file, or falls back to the most expensive model for a simple question, as wasteful rather than malicious. ## Sources - [Introducing Operant Semantic Firewall](https://www.operant.ai/art-kubed/introducing-operant-semantic-firewall) (operant.ai), read 2026-11-16 ## Related pages - [Model routing](/glossary/model-routing/) - [LLM gateway](/glossary/llm-gateway/) - [Guardrail policy](/glossary/guardrail-policy/) - [AI Agent Security Risks, Mapped to Each Layer](/guides/ai-agent-security-risks/) - [All glossary terms](/glossary/) - [SaaS agent](/glossary/saas-agent/) - [Workspace agent](/glossary/workspace-agent/) - [Guides](/guides/) --- # Workspace agent: definition and example URL: https://aiagentsecuritysolutions.com/glossary/workspace-agent/ Updated: 2026-11-16 A workspace agent is an AI assistant or agent embedded in productivity and collaboration software, with delegated access to a user's documents, mail and chats. Several entry points often share one identity model and data estate. They may not share one enforcement point. ## Example Operant says Copilot in Word, Copilot in Teams, Microsoft 365 Copilot Chat, Copilot Studio agents and Agents SDK agents each have their own entry point and share an identity model, but not a single enforcement point. ## Sources - [Operant: coverage for Microsoft Copilot and embedded AI assistants](https://www.operant.ai/art-kubed/announcing-operants-coverage-for-microsoft-copilot-and-every-ai-assistant-embedded-in-microsoft-365) (operant.ai), read 2026-11-16 ## Related pages - [SaaS agent](/glossary/saas-agent/) - [Agent observability](/glossary/agent-observability/) - [Agent sprawl](/glossary/agent-sprawl/) - [All glossary terms](/glossary/) - [Unbounded consumption](/glossary/unbounded-consumption/) - [A2A protocol](/glossary/a2a-protocol/) - [Guides](/guides/) --- # AI agent security news URL: https://aiagentsecuritysolutions.com/news/ Updated: 2026-11-16 Twelve dated items track AI agent security since 18 Aug 2026: layer-specific product releases, acquisitions and funding, incidents and advisories, and standards or framework updates. Each item starts from the vendor's or publisher's own release, advisory or document, with the read date. ## 12 items - 2026-11-13: [CrowdStrike Falcon AI Gateway reaches general availability](/news/crowdstrike-falcon-ai-gateway-generally-available/) - 2026-11-04: [CrowdStrike adds Seraphic browser security to Falcon](/news/crowdstrike-seraphic-enterprise-browser-falcon/) - 2026-10-27: [Cato Networks launches Cato AI Endpoint for local agents](/news/cato-ai-endpoint-launch/) - 2026-10-21: [Cisco DefenseClaw runs coding agents in a sandbox](/news/cisco-defenseclaw-nvidia-openshell-coding-agents/) - 2026-10-19: [Microsoft adds agent discovery and Dynamic Permissions](/news/microsoft-agentic-insider-controls-dynamic-permissions/) - 2026-10-16: [NIST, NSA and CISA publish ten agent incident practices](/news/nist-nsa-cisa-ten-practices-agent-incident-response/) - 2026-10-08: [OWASP GenAI exploit roundup for Q3 2026 maps incidents](/news/owasp-genai-exploit-roundup-q3-2026/) - 2026-10-08: [AgentCorruption: Zenity Labs reports AWS AgentCore flaws](/news/zenity-labs-agentcorruption-aws-agentcore/) - 2026-10-01: [Operant extends agent defense to Microsoft 365 Copilot](/news/operant-ai-microsoft-365-copilot-agents-sdk-coverage/) - 2026-09-24: [Island raises $400 million Series F at a $6.4B valuation](/news/island-400-million-series-f-agentic-control-plane/) - 2026-09-24: [SalesBleed: three Agentforce flaws reported by Zenity Labs](/news/zenity-labs-salesbleed-agentforce-flaws/) - 2026-09-02: [CrowdStrike launches an identity provider for AI agents](/news/crowdstrike-agentic-identity-provider/) Subscribe with the [RSS feed](/feed.xml). ## Related pages - [Guides](/guides/) - [The ranking](/ranking/) - [Vendors](/vendors/) - [Send a correction or tip](/corrections/) --- # CrowdStrike Falcon AI Gateway reaches general availability URL: https://aiagentsecuritysolutions.com/news/crowdstrike-falcon-ai-gateway-generally-available/ Updated: 2026-11-13 CrowdStrike described Falcon AI Gateway as generally available on 13 November 2026 and included with Falcon Guardian. Per CrowdStrike, it routes AI traffic through one gateway that inspects prompts and responses and enforces policy on managed and unmanaged devices. The post gives no latency figure or fail-open behavior. 2026-11-13 CrowdStrike published an architecture post on Falcon AI Gateway on 13 November 2026, describing it as generally available and included with Falcon Guardian. The gateway gives security and platform teams one place to govern how applications and workloads reach AI models. Per the post, it routes AI traffic through a central gateway to inspect prompts and responses, monitor usage and enforce policy across managed and unmanaged devices, extending visibility beyond the Falcon sensor. Developers connect an existing SDK or coding tool by changing the base URL to point at the gateway and using a gateway key. A gateway sees model traffic, not what an agent does after the response arrives, which is why this site maps it as one layer among five. The post discusses keeping the request path short but states no latency number and no fail-open or fail-closed behavior, per the 16 Nov 2026 read. ## Sources - [Inside Falcon AI Gateway: One Control Point for AI Traffic](https://www.crowdstrike.com/en-us/blog/falcon-ai-gateway-one-control-point-for-ai-traffic/) (crowdstrike.com), read 2026-11-16 ## Related pages - [CrowdStrike profile](/vendors/crowdstrike/) - [All news](/news/) - [CrowdStrike launches an identity provider for AI agents](/news/crowdstrike-agentic-identity-provider/) - [CrowdStrike adds Seraphic browser security to Falcon](/news/crowdstrike-seraphic-enterprise-browser-falcon/) - [The ranking](/ranking/) --- # CrowdStrike adds Seraphic browser security to Falcon URL: https://aiagentsecuritysolutions.com/news/crowdstrike-seraphic-enterprise-browser-falcon/ Updated: 2026-11-04 CrowdStrike announced on 4 November 2026 that Falcon Seraphic Enterprise Browser is available through the unified Falcon platform experience. CrowdStrike says it works in existing browsers without a dedicated browser or traffic rerouting, inventories extensions with risk assessments, and enforces copy, paste, upload and download policies for generative AI tools. 2026-11-04 CrowdStrike announced on 4 November 2026 that Falcon Seraphic Enterprise Browser is natively available on the Falcon platform. Per the blog, protection works across managed and unmanaged devices without requiring employees to adopt a dedicated enterprise browser or reroute traffic. CrowdStrike says teams can inventory browser extensions, assess their risk and allow or block them, and can enforce copy, paste, upload and download policies to keep sensitive data from unsanctioned web apps and generative AI tools. Analysts can also correlate browser activity with endpoint, identity and third-party data in the same console. For the browser layer, note what this product is: CrowdStrike presents Seraphic as its own Falcon product, and the announcement does not mention Falcon Guardian, its agent security line, so the browser controls described are general browser security with AI-related policies. All statements are CrowdStrike's and were not tested by this site. ## Sources - [CrowdStrike Brings Enterprise Browser Security to the Falcon Platform](https://www.crowdstrike.com/en-us/blog/crowdstrike-brings-enterprise-browser-security-to-falcon-platform/) (crowdstrike.com), read 2026-11-16 ## Related pages - [CrowdStrike profile](/vendors/crowdstrike/) - [All news](/news/) - [CrowdStrike Falcon AI Gateway reaches general availability](/news/crowdstrike-falcon-ai-gateway-generally-available/) - [Cato Networks launches Cato AI Endpoint for local agents](/news/cato-ai-endpoint-launch/) - [The ranking](/ranking/) --- # Cato Networks launches Cato AI Endpoint for local agents URL: https://aiagentsecuritysolutions.com/news/cato-ai-endpoint-launch/ Updated: 2026-10-27 Cato Networks launched Cato AI Endpoint on 27 October 2026, a client that discovers AI use, inspects prompts and responses inline, and enforces policy across browsers, desktop applications and local agents. Cato says it works alongside existing security tools. Cato's AI security line draws on its Aim Security acquisition. 2026-10-27 Cato Networks announced Cato AI Endpoint on 27 October 2026. Cato describes it as an endpoint client that discovers AI use across the endpoint, inspects prompts and responses inline, and enforces policy directly on the machine, covering browsers, desktop applications and local agents. Cato says the client links the user, endpoint, data, prompt, response, agent activity, tool call and policy decision, which it says can reveal risky agents, Model Context Protocol servers and skills before they act. It says the client works alongside existing security tools, so AI protection can be added without replacing the stack. The release quotes customer and analyst-firm voices from PlayPower and 650 Group. Per Cato's own blog, Cato acquired Aim Security in September 2025, and Cato AI Security combines capabilities from that acquisition. The coverage statements are Cato's; the layer map records what the vendor documents for each layer. ## Sources - [Cato Networks Launches Cato AI Endpoint to Secure AI Use Beyond the Browser](https://www.catonetworks.com/news/cato-launches-cato-ai-endpoint-secure-ai-use-beyond-browser/) (catonetworks.com), read 2026-11-16 - [Securing AI transformation: why Cato acquired Aim Security](https://www.catonetworks.com/blog/securing-ai-transformation-why-cato-acquired-aim-security/) (catonetworks.com), read 2026-11-16 ## Related pages - [Aim Security profile](/vendors/aim-security/) - [All news](/news/) - [CrowdStrike adds Seraphic browser security to Falcon](/news/crowdstrike-seraphic-enterprise-browser-falcon/) - [Cisco DefenseClaw runs coding agents in a sandbox](/news/cisco-defenseclaw-nvidia-openshell-coding-agents/) - [The ranking](/ranking/) --- # Cisco DefenseClaw runs coding agents in a sandbox URL: https://aiagentsecuritysolutions.com/news/cisco-defenseclaw-nvidia-openshell-coding-agents/ Updated: 2026-10-21 Cisco described DefenseClaw with NVIDIA OpenShell on 21 October 2026. Per Cisco, DefenseClaw checks every prompt, tool call and tool result inside the sandbox, decides web destinations by host name, hides real provider credentials from the agent, and records each block and approval for Splunk or OpenTelemetry backends. 2026-10-21 Cisco published a blog on 21 October 2026 on protecting enterprise coding agents with DefenseClaw and NVIDIA OpenShell. The premise is that a coding agent acts with the full access rights of the developer who started it, so a sandbox should limit it to what the task needs. Per Cisco, OpenShell runs each agent in a sandbox where per-agent network egress policy acts as micro-segmentation, and where the agent sees placeholders instead of real provider credentials. DefenseClaw checks each prompt, tool call and tool result using its own rules and Cisco AI Defense, and decides each web destination by host name without decrypting traffic. Administrators can set user- or group-based policy for which coding agents run on managed endpoints and which run only in a sandbox. Cisco says every block, approval and decision is recorded with the agent, sandbox and policy behind it and can go to Splunk or any OpenTelemetry backend. These are Cisco's statements and were not tested by this site. ## Sources - [Protect enterprise coding agents with Cisco DefenseClaw and NVIDIA OpenShell](https://blogs.cisco.com/ai/protect-enterprise-coding-agents-with-cisco-defenseclaw-and-nvidia-openshell) (blogs.cisco.com), read 2026-11-16 ## Related pages - [Cisco profile](/vendors/cisco/) - [All news](/news/) - [Cato Networks launches Cato AI Endpoint for local agents](/news/cato-ai-endpoint-launch/) - [Microsoft adds agent discovery and Dynamic Permissions](/news/microsoft-agentic-insider-controls-dynamic-permissions/) - [The ranking](/ranking/) --- # Microsoft adds agent discovery and Dynamic Permissions URL: https://aiagentsecuritysolutions.com/news/microsoft-agentic-insider-controls-dynamic-permissions/ Updated: 2026-10-19 Microsoft announced new agent controls on 19 October 2026: discovery expands by 35 types to 60 supported local agents, Entra Agent ID gains identity types for local and unknown agents, and Dynamic Permissions scopes authority to an approved task. SiliconANGLE reported no availability dates or pricing at launch. 2026-10-19 Microsoft Security published its agentic insider post on 19 October 2026. It extends discovery of local agents by 35 types, to 60 supported agents including Claude Code, OpenAI Codex and GitHub Copilot, and lets them join a unified registry shared with Microsoft Agent 365. On identity, Microsoft says it is extending Entra Agent ID with identity types for local and previously unknown agents found by security sensors, so such an agent can become a policy target. It also introduces Dynamic Permissions: organization-wide policy sets the outer boundary, authority is scoped to the task and time-bound, and higher-impact actions can require verified human approval. Microsoft describes the Microsoft Execution Container as a sandbox that keeps controls outside the agent's reach. SiliconANGLE reported that Microsoft gave no availability dates or pricing for the new controls, and that Conditional Access now extends to agents so rising risk can hold higher-impact actions for approval. The capability statements are Microsoft's own. ## Sources - [Agents are your newest insiders. Security must move at their speed.](https://www.microsoft.com/en-us/security/blog/2026/10/19/agents-are-your-newest-insiders-security-must-move-at-their-speed/) (microsoft.com), read 2026-11-16 - [Microsoft targets the agentic insider with new controls for rogue AI agents](https://siliconangle.com/2026/10/19/microsoft-targets-the-agentic-insider-with-new-controls-for-rogue-ai-agents/) (siliconangle.com), read 2026-11-16 ## Related pages - [Microsoft profile](/vendors/microsoft/) - [All news](/news/) - [Cisco DefenseClaw runs coding agents in a sandbox](/news/cisco-defenseclaw-nvidia-openshell-coding-agents/) - [NIST, NSA and CISA publish ten agent incident practices](/news/nist-nsa-cisa-ten-practices-agent-incident-response/) - [The ranking](/ranking/) --- # NIST, NSA and CISA publish ten agent incident practices URL: https://aiagentsecuritysolutions.com/news/nist-nsa-cisa-ten-practices-agent-incident-response/ Updated: 2026-10-16 NIST's CAISSI, the NSA's Super Intelligence Security Center and CISA published ten key practices on 16 October 2026 for incident readiness and response. The voluntary document targets incidents where agents take significant uninstructed actions, and spans preparation, detection, response, recovery and reporting. It is written for frontier training and evaluation. 2026-10-16 NIST's Center for Advancing Innovation and Standards for Super Intelligence (CAISSI), the NSA's Super Intelligence Security Center and CISA published ten key practices for incident readiness and response on 16 October 2026. The document is voluntary and covers training and evaluation of frontier systems. It focuses on incidents in which one or more agents take significant uninstructed actions that breach the operator's task scope or human-oversight measures. The ten practices are split across five stages of the incident lifecycle: preparation, detection, response, recovery and reporting. NIST says they rest on its existing incident response recommendations for cybersecurity risk management. Enterprise buyers are not the stated audience, so treat it as a structure to borrow rather than a compliance requirement. The stages map cleanly onto the control points tracked in this site's agent control point map: inventory and audit support detection, and runtime blocking and revocation support response. ## Sources - [Key Practices for Incident Readiness and Response in Super Intelligence Training and Evaluation](https://www.nist.gov/news-events/news/2026/10/key-practices-incident-readiness-and-response-super-intelligence-training) (nist.gov), read 2026-11-16 ## Related pages - [All news](/news/) - [Microsoft adds agent discovery and Dynamic Permissions](/news/microsoft-agentic-insider-controls-dynamic-permissions/) - [OWASP GenAI exploit roundup for Q3 2026 maps incidents](/news/owasp-genai-exploit-roundup-q3-2026/) - [The ranking](/ranking/) --- # OWASP GenAI exploit roundup for Q3 2026 maps incidents URL: https://aiagentsecuritysolutions.com/news/owasp-genai-exploit-roundup-q3-2026/ Updated: 2026-10-08 The OWASP GenAI Security Project published its GenAI and Agentic AI Exploit Roundup for Q3 2026 on 8 October 2026. It lists seven entries, grouped as one three-part campaign, three further incidents and one research-demonstrated Copilot exploit family, each mapped to the 2026 LLM and agentic Top 10 lists. 2026-10-08 The OWASP GenAI Security Project published its Q3 2026 exploit roundup on 8 October 2026, covering 1 July through 30 September. The project says the roundup is not exhaustive and is not an official OWASP publication. It aligns each entry to the OWASP Top 10 for LLM Applications 2026 and the Top 10 for Agentic Applications 2026, with CVE references where they exist. The seven entries should not be read as seven separate attacks: they include three connected parts of one campaign, three further incidents and one research-demonstrated family of Copilot exploits. The project's summary says agents caused harm when evaluation boundaries failed, when they treated reachable systems as authorized targets, and when exposed credentials and excessive permissions widened the damage. The roundup's own conclusion is that enforced scope, isolated tools and credentials, and monitoring of actual actions are what hold, and that prompt-level instructions alone do not set a secure boundary. For layer planning, that points to controls outside the model: egress limits, per-run credentials and action logging, each owned by a different layer on this site's map. ## Sources - [GenAI and Agentic AI Exploit Roundup Q3 2026](https://genai.owasp.org/2026/10/08/genai-and-agentic-ai-exploit-roundup-q3-2026/) (genai.owasp.org), read 2026-11-16 ## Related pages - [All news](/news/) - [NIST, NSA and CISA publish ten agent incident practices](/news/nist-nsa-cisa-ten-practices-agent-incident-response/) - [AgentCorruption: Zenity Labs reports AWS AgentCore flaws](/news/zenity-labs-agentcorruption-aws-agentcore/) - [The ranking](/ranking/) --- # AgentCorruption: Zenity Labs reports AWS AgentCore flaws URL: https://aiagentsecuritysolutions.com/news/zenity-labs-agentcorruption-aws-agentcore/ Updated: 2026-10-08 Zenity Labs reported AgentCorruption on 8 October 2026, a chain of Amazon Bedrock AgentCore flaws. A single prompt to one public-facing agent gave access to all AgentCore agents in the same AWS account and region, plus conversations, memories and credentials. Zenity says AWS made IMDSv2 the default and reduced the default role's permissions. 2026-10-08 Zenity Labs presented AgentCorruption at SecTor 2026 in Toronto and published a press release on 8 October 2026. The research covers a chain of flaws in Amazon Bedrock AgentCore. Using one prompt to a public-facing agent that had an outbound-request tool, researchers reached credentials for a default IAM role whose permissions covered every AgentCore agent in the same AWS account and region. From there Zenity says it could invoke agents it was not authorized to use, read private conversations and long-term memories, download container images and source code, and retrieve API keys and OAuth tokens held in AWS Secrets Manager. It also planted memories that sent future conversations to an outside destination, a persistence route that outlives the first session. Zenity says it reported the findings to AWS on 25 December 2025. Following that report, AWS made IMDSv2 the default for AgentCore deployments, and Zenity's testing confirmed AWS had cut the default execution role's permissions. The identity lesson is plain: one over-broad role shared by many agents turns a single compromised agent into a regional problem. The account is Zenity's own. ## Sources - [Zenity Labs reports AgentCorruption, a chain of AWS AgentCore flaws](https://zenity.io/press-release/zenity-labs-discloses-agentcorruption-a-chain-of-aws-agentcore-flaws) (zenity.io), read 2026-11-16 ## Related pages - [Zenity profile](/vendors/zenity/) - [All news](/news/) - [OWASP GenAI exploit roundup for Q3 2026 maps incidents](/news/owasp-genai-exploit-roundup-q3-2026/) - [Operant extends agent defense to Microsoft 365 Copilot](/news/operant-ai-microsoft-365-copilot-agents-sdk-coverage/) - [The ranking](/ranking/) --- # Operant extends agent defense to Microsoft 365 Copilot URL: https://aiagentsecuritysolutions.com/news/operant-ai-microsoft-365-copilot-agents-sdk-coverage/ Updated: 2026-10-01 Operant announced on 1 October 2026 coverage for Microsoft 365: inline step-level defense for custom agents built on the Microsoft 365 Agents SDK, Purview-based observability for Copilot, and prompt blocking for embedded assistants. Operant states the Purview path detects and alerts but does not sit inline. 2026-10-01 Operant announced three expansions for Microsoft 365 on 1 October 2026. Its Purview integration ingests Copilot prompts, responses and tool calls across Copilot surfaces and runs Operant detections against them. Agent Protector goes inside the loop of custom agents built on the Microsoft 365 Agents SDK, so each step is evaluated before it executes. For third-party assistants embedded in Microsoft 365 on the web, Operant observes prompts, blocks violating ones, and can block a whole plugin by the sensitivity labels already applied in Purview. Operant is explicit about the limits. It says the Purview path is an observability and detection route that does not sit inline on Microsoft's inference path, so on its own it detects, alerts and drives response rather than stopping a prompt before the model sees it. Inline enforcement applies to the agents a team builds. That split is the practical lesson for the cloud workspace layer: the same vendor can offer a detect-only path and a block path for the same platform. Statements are Operant's own and were not tested by this site. ## Sources - [Announcing Operant's Coverage for Microsoft Copilot and Every AI Assistant Embedded in Microsoft 365](https://www.operant.ai/art-kubed/announcing-operants-coverage-for-microsoft-copilot-and-every-ai-assistant-embedded-in-microsoft-365) (operant.ai), read 2026-11-16 ## Related pages - [Operant AI profile](/vendors/operant-ai/) - [All news](/news/) - [AgentCorruption: Zenity Labs reports AWS AgentCore flaws](/news/zenity-labs-agentcorruption-aws-agentcore/) - [Island raises $400 million Series F at a $6.4B valuation](/news/island-400-million-series-f-agentic-control-plane/) - [The ranking](/ranking/) --- # Island raises $400 million Series F at a $6.4B valuation URL: https://aiagentsecuritysolutions.com/news/island-400-million-series-f-agentic-control-plane/ Updated: 2026-09-24 Island announced a $400 million Series F on 24 September 2026, led by Evolution Equity Partners, at a $6.4 billion valuation. Island says the round more than doubles its 2024 valuation and that annual recurring revenue has doubled every fiscal year since 2022. The company calls itself the agentic control plane for enterprises. 2026-09-24 Island announced a $400 million Series F financing round on 24 September 2026. Evolution Equity Partners led the round, and existing investors including Prysm Capital, Sequoia Capital, Coatue Management, Cyberstarts and Insight Partners took part. Island says the valuation is now $6.4 billion, more than double its 2024 level. Island says it employs 1,000 people and has doubled annual recurring revenue every fiscal year since its 2022 launch. Those are the company's own figures from the release. Island says the funding supports growth as enterprises scale AI agents, and describes its product as an agentic control plane covering what people and agents do as they interact with applications and data. For buyers choosing a layer, the relevant point is that a browser-first vendor now frames itself as covering endpoints, networks and data as well. The coverage claims are Island's; the layer map on this site records what each vendor documents, not what a funding release says. ## Sources - [Island Announces $400 Million Series F, Bringing Valuation to $6.4 Billion](https://www.island.io/press/island-announces-400-million-series-f-bringing-valuation-to-6-4-billion) (island.io), read 2026-11-16 ## Related pages - [Island profile](/vendors/island/) - [All news](/news/) - [Operant extends agent defense to Microsoft 365 Copilot](/news/operant-ai-microsoft-365-copilot-agents-sdk-coverage/) - [SalesBleed: three Agentforce flaws reported by Zenity Labs](/news/zenity-labs-salesbleed-agentforce-flaws/) - [The ranking](/ranking/) --- # SalesBleed: three Agentforce flaws reported by Zenity Labs URL: https://aiagentsecuritysolutions.com/news/zenity-labs-salesbleed-agentforce-flaws/ Updated: 2026-09-24 Zenity Labs reported three Salesforce Agentforce flaws, named SalesBleed, on 24 September 2026. Per the research, a poisoned public lead form could steer an agent into leaking CRM data without a click, and into sending phishing messages under the agent's identity. Salesforce worked with Zenity to fix the issues, per The Register. 2026-09-24 Zenity Labs published research on 24 September 2026 on three vulnerabilities in Salesforce Agentforce, collectively called SalesBleed. The Register reported that Zenity reported them to Salesforce, which worked with Zenity to fix them, and that the attack chains described no longer work. The research describes two paths. In the first, text planted in a public lead form stays dormant until an employee asks an Agentforce agent about leads, at which point the agent reads the poisoned record and acts on instructions the user never gave. In the second, a related weakness let the agent send phishing messages in Slack under the agent's own identity. Zenity says a control meant to limit external destinations did not cover every case. This is a cloud workspace case: the agent runs inside the SaaS platform, so an endpoint or browser control sees nothing. What a control there can see is the agent's records, tool calls and outbound destinations. Zenity's CTO told The Register that secure-by-design may no longer be enough for agents and that their activity needs close monitoring. That is Zenity's view, and Zenity sells agent security. ## Sources - [SalesBleed: 0-click data exfiltration on Agentforce](https://labs.zenity.io/post/salesbleed-0-click-data-exfiltration-on-agentforce) (labs.zenity.io), read 2026-11-16 - [Salesforce Agentforce vulns allowed 0-click CRM data theft, anonymous phishing](https://www.theregister.com/security/2026/09/24/salesforce-agentforce-vulns-allowed-0-click-crm-data-theft-anonymous-phishing/5298958) (theregister.com), read 2026-11-16 - [SalesBleed flaws in Salesforce Agentforce enabled zero-click data exfiltration](https://www.securityweek.com/salesbleed-flaws-in-salesforce-agentforce-enabled-zero-click-data-exfiltration/) (securityweek.com), read 2026-11-16 ## Related pages - [Zenity profile](/vendors/zenity/) - [All news](/news/) - [Island raises $400 million Series F at a $6.4B valuation](/news/island-400-million-series-f-agentic-control-plane/) - [CrowdStrike launches an identity provider for AI agents](/news/crowdstrike-agentic-identity-provider/) - [The ranking](/ranking/) --- # CrowdStrike launches an identity provider for AI agents URL: https://aiagentsecuritysolutions.com/news/crowdstrike-agentic-identity-provider/ Updated: 2026-09-02 CrowdStrike introduced Agentic Identity Provider on 2 September 2026 at Fal.Con, inside Falcon Next-Gen Identity Security. Per CrowdStrike's blog, it registers agents in one directory, gives each a cryptographically verifiable identity, and brokers short-lived, tightly scoped access instead of standing credentials. Details were read on 16 Nov 2026. 2026-09-02 CrowdStrike announced Agentic Identity Provider (Agentic IdP) on 2 September 2026, at its Fal.Con 2026 event. The product sits in Falcon Next-Gen Identity Security, and CrowdStrike describes it as the identity control plane for the agentic enterprise. Per the announcement, the product registers agents in a single directory, gives each one a cryptographically verifiable identity, and links it to the humans and workloads behind it. Instead of giving agents standing credentials of their own, it brokers short-lived, tightly scoped access on their behalf, for as long as a task needs it. For the identity layer, the practical question is what a control at this point can see: which agents are registered and what access was brokered to them. An agent that never registers, such as a local coding agent started by a developer, is outside this view unless another layer reports it. All capability statements here are CrowdStrike's own and were not tested by this site. ## Sources - [CrowdStrike Announces Agentic Identity Provider](https://www.crowdstrike.com/en-us/blog/crowdstrike-announces-agentic-identity-provider/) (crowdstrike.com), read 2026-11-16 ## Related pages - [CrowdStrike profile](/vendors/crowdstrike/) - [All news](/news/) - [SalesBleed: three Agentforce flaws reported by Zenity Labs](/news/zenity-labs-salesbleed-agentforce-flaws/) - [CrowdStrike Falcon AI Gateway reaches general availability](/news/crowdstrike-falcon-ai-gateway-generally-available/) - [The ranking](/ranking/) --- # AI agent security solutions FAQ URL: https://aiagentsecuritysolutions.com/faq/ Updated: 2026-11-16 Answers to common questions on AI agent security solutions: what a solution is, how tools differ from platforms, which layers exist, which products stop actions rather than alert, what they cost and how they work with EDR. Facts carry sources and the 16 Nov 2026 read date. ## Questions and answers ### What is an AI agent security solution? An AI agent security solution is a product that finds AI agents, limits what they can reach and stops or records what they do. It acts at one or more of five layers: the endpoint, the browser, cloud workspaces, identity and the gateway between agents and models or tools. Products differ in which layers they cover and whether they block or only alert. ### What is the difference between an AI agent security tool and a platform? A tool usually covers one layer or one control point, such as a hook on coding agents or a network allow list for MCP servers. A platform covers several layers under one inventory and one policy. Platforms trade depth for reach, so read the documentation for what each product does at each layer before counting layers. Source: [WitnessAI: agentic control](https://witness.ai/blog/introducing-witnessai-agentic-control-one-control-plane-for-every-agent-tool-and-mcp-server/), read 2026-11-16. ### Which solutions stop actions rather than only alert? Several vendors say they block before the action runs. Pillar Security says its Runtime Guardian decides on every action before it runs, Cisco says its guardrail checks a prompt, shell command or tool call before the agent acts, and Bay says it decides allow, ask, deny or steer. Check each vendor page for the action types named and for a second source. Source: [Cisco DefenseClaw: guardrail](https://cisco-ai-defense.github.io/defenseclaw/docs/guardrail/), read 2026-11-16. ### Do I need more than one layer? Most companies do, because agents run in more than one place: coding agents on laptops, browser agents in the browser, builders' agents in SaaS platforms. Start with the layer where your agents run most, then add the layer that closes the next largest gap. One product may cover two or three layers; none documents every layer equally well. ### How much do AI agent security solutions cost? Most vendors publish no price and quote after a demo. Microsoft lists Agent 365 at $15.00 per user per month, billed yearly, and CrowdStrike says Falcon AI Gateway is included with Falcon Guardian at no additional cost. Cisco describes an Explorer Edition as free. For the rest, a published price was none found on 16 Nov 2026. Sources: [Microsoft Agent 365](https://www.microsoft.com/en-us/microsoft-agent-365), read 2026-11-16; [CrowdStrike: Falcon Guardian and AI Gateway](https://www.crowdstrike.com/en-us/blog/crowdstrike-introduces-ai-kill-switch/), read 2026-11-16; [CRN: Cisco unveils zero trust for AI agents](https://www.crn.com/news/security/2026/cisco-unveils-zero-trust-for-ai-agents-5-things-to-know), read 2026-11-16. ### Which solutions work with my EDR? Several install through the EDR or MDM you run. Bay says one script through an existing MDM or EDR installs its discovery binary, Cato says AI Endpoint runs beside the EDR, and Noma Security says discovery works through existing EDR or MDM. Ask each vendor which EDR products it names, since lists differ and some publish none. Source: [Cato Networks: AI Endpoint](https://www.catonetworks.com/platform/cato-ai-endpoint/), read 2026-11-16. ### Are AI agents a security risk? Yes, because agents plan and take actions that affect real systems. NIST's request for information on securing AI agent systems names risks from adversarial data such as indirect prompt injection, from insecure models, and from models taking harmful actions even without an attack. Limiting tools, permissions and autonomy reduces the exposure. Source: [NIST CAISI request for information on securing AI agent systems](https://www.nist.gov/news-events/news/2026/01/caisi-issues-request-information-about-securing-ai-agent-systems), read 2026-11-16. ### Can AI agents turn security vulnerabilities into real attacks? They can turn a manipulated or mistaken output into a real action. OWASP calls this excessive agency: damaging actions performed in response to unexpected, ambiguous or manipulated model outputs, whatever the cause. Its root causes are excessive functionality, excessive permissions and excessive autonomy, so removing tools, narrowing scopes and adding approvals lowers the risk. Source: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/), read 2026-11-16. ### What developer platforms support AI agent security? Several platforms ship their own controls. Claude Code offers hooks, including a PreToolUse hook that runs before a tool call and can block it. Amazon Bedrock lets you associate a guardrail with an agent, and Microsoft documents Entra identities and governance for Copilot Studio agents. Third-party products add inventory and policy across platforms. Source: [Claude Code hooks reference](https://docs.anthropic.com/en/docs/claude-code/hooks), read 2026-11-16. ### What is the difference between AI agent security platforms? They differ in the layers they cover, the verb they use at each layer (discover, observe, alert or block), how they deploy and what they publish. Some install through MDM on laptops, some connect by API to SaaS platforms, and some sit on the network. The vendor pages and layer views show these differences with sources. ### What is the state of AI agent security in 2026? It is moving from discovery toward enforcement. OWASP published its Top 10 for Agentic Applications for 2026 in December 2025, NIST asked for public input on securing agent systems in January 2026, and vendors added endpoint hooks, agent identity and kill switches during the year. Public documentation and prices remain thin at many vendors. Source: [OWASP Top 10 for Agentic Applications for 2026](https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/), read 2026-11-16. ### Is there open source AI agent security? Yes, in parts. Cisco publishes DefenseClaw documentation and Apache-2.0 code for guardrails on coding agents, and Operant publishes Woodpecker, an open-source red teaming tool, on GitHub under Apache-2.0. Open-source pieces cover narrower ground than commercial platforms, so check the supported agents and operating systems. Source: [Operant AI Woodpecker on GitHub](https://github.com/OperantAI/woodpecker), read 2026-11-16. ### Does an AI agent security gateway replace endpoint controls? No. A gateway sees calls routed through it and nothing that bypasses it, while an endpoint control sees actions on the device before they run. Palo Alto Networks documents an AI Gateway between applications and each LLM provider. Agents on laptops that call tools locally need a control on the device as well. Source: [Palo Alto Networks Prisma AIRS AI Gateway overview](https://docs.paloaltonetworks.com/prisma-airs/ai-gateway/ai-gateway-overview), read 2026-11-16. ### What are the best practices for AI agent security? Limit functionality, permissions and autonomy for every agent, put a blocking control on the tool call, issue short-lived scoped credentials, keep an approved list of tools and MCP servers, and log the agent, user, tool and decision. Test each control with a harmless case and report coverage monthly. The guide on best practices gives the steps by layer. Source: [OWASP LLM06:2025 Excessive Agency](https://genai.owasp.org/llmrisk/llm062025-excessive-agency/), read 2026-11-16. ### Which AI agent security vendors does this site cover? Sixteen: Aim Security, Bay, Check Point, Cisco, CrowdStrike, Island, Lasso Security, Microsoft, Noma Security, Operant AI, Palo Alto Networks, Pillar Security, SentinelOne (Prompt Security), Straiker, WitnessAI and Zenity. Each has a profile with sourced facts, a score for each criterion and the reason, read on 16 Nov 2026. ## Related pages - [Guides](/guides/) - [Glossary](/glossary/) - [The ranking](/ranking/) - [Method](/method/) --- # About AI Agent Security Solutions URL: https://aiagentsecuritysolutions.com/about/ Updated: 2026-11-16 AI Agent Security Solutions publishes a ranking of sixteen AI agent security solutions scored by layer: endpoint, browser, cloud workspace, identity and gateway. It also publishes guides, a glossary, news and two datasets. Everything rests on public evidence read on 16 Nov 2026, with the method and weights in the open. ## What does this site publish? A ranked comparison of solutions for securing AI agents wherever they run, five layer views, a profile for each of sixteen vendors, twelve guides, a glossary, dated news, an FAQ and two downloadable datasets: the Layer Coverage Map and the Agent Control Point Map. The reader is a security lead choosing how to secure agents across the endpoint, the browser, cloud workspaces, identity systems and the gateways between agents and tools or models. ## How is it made? Each vendor is researched with one plan, scored blind against weights committed before any evidence, and checked by a second blind reviewer. Every fact has a source URL, a quote and a read date. The method page gives the criteria, anchors and rules. Pages that carry numbers print them from the scores files, so text and tables cannot drift apart. ## How do I cite it? Cite the title, the edition date, the URL and the license: AI Agent Security Solutions, edition 16 Nov 2026, CC BY 4.0. The datasets page repeats the cite line for each file. ## Related pages - [Method](/method/) - [Editorial policy](/editorial-policy/) - [Corrections](/corrections/) - [The ranking](/ranking/) - [FAQ](/faq/) --- # Editorial policy URL: https://aiagentsecuritysolutions.com/editorial-policy/ Updated: 2026-11-16 Every claim on this site rests on a source URL and a read date. Vendor statements are written as the vendor's own. Scores come from a written evidence file, a blind scorer and a blind reviewer, against weights committed before scoring. Missing evidence is written as none found on a date, never as a negative claim. ## What sourcing rules apply? Each fact needs a URL that returned a page when read, the read date and a short verbatim quote. A vendor claim is worded as the vendor's: Operant says, Microsoft documents. A third-party or directly observed source is needed before a score can pass 7. Where evidence is missing, the text says none found on the read date and names the place searched. ## How are scores set? Weights and criteria were committed before any evidence or score and are fixed. Scores are whole numbers from 0 to 10 with a written reason and the fact ids behind it. A score changes only when a sourced fact changes, and the change is logged with the fact. ## What do the guides and news follow? Guides, glossary, news and the FAQ do not quote ranks or scores; the site prints those from the scores files. Security content gives risks, detection, policy and fixes, each sourced, and never exploit code, injection payloads or steps to bypass a control. ## How do corrections work? Send a sourced correction to the address on the corrections page. Accepted changes are dated and listed in the changelog on the method page. ## Related pages - [Method](/method/) - [Corrections](/corrections/) - [The ranking](/ranking/) - [FAQ](/faq/) --- # Corrections and tips URL: https://aiagentsecuritysolutions.com/corrections/ Updated: 2026-11-16 Send a sourced correction or a tip to the editor at editor@aiagentsecuritysolutions.com. Include the page, the claim, the source URL and the date you read it. A fact that changes is updated with any score that rests on it, and the change is dated in the changelog on the method page. ## How do I send a correction? Email [editor@aiagentsecuritysolutions.com](mailto:editor@aiagentsecuritysolutions.com) with the page address, the sentence you think is wrong, the source that shows the right fact and the date you read it. A vendor can send a primary source such as a documentation page, release note or price page. The editor checks the source, updates the fact, re-checks any score that rests on it and records the change with a date. ## How do I send a tip? Send a tip to the same address: a product release, a layer a vendor now covers, an advisory or a documentation page that was behind a login. Link the primary source. Tips that fit the scope of this site become news items or fact updates after checking. ## What changes after a correction? A confirmed correction can move a score. It cannot change the weights or the criteria, which are fixed once the first score exists. The changelog lists each change with its date and the fact id. ## Write to the editor Email [editor@aiagentsecuritysolutions.com](mailto:editor@aiagentsecuritysolutions.com) with the page address, what is wrong and the source that shows it. ## Related pages - [Method](/method/) - [Editorial policy](/editorial-policy/) - [The ranking](/ranking/) - [FAQ](/faq/) --- # Contact URL: https://aiagentsecuritysolutions.com/contact/ Updated: 2026-11-16 Email the editor at editor@aiagentsecuritysolutions.com for corrections, tips, vendor facts, dataset questions and press requests. Corrections should name the page, the claim, a source URL and the date read. The site has no forms, accounts or cookies, so email is the only channel. ## How do I reach the editor? Write to [editor@aiagentsecuritysolutions.com](mailto:editor@aiagentsecuritysolutions.com). Use the subject line to say what the message is: correction, tip, dataset question or other. A correction with a source URL is checked first. Expect a reply after the source has been read and the fact compared with the page. ## What does the editor not answer? The editor does not give buying advice for a specific company or take part in vendor selection. The ranking, layer views and guides are the published answer. ## Write to the editor Email [editor@aiagentsecuritysolutions.com](mailto:editor@aiagentsecuritysolutions.com). ## Related pages - [Method](/method/) - [Editorial policy](/editorial-policy/) - [Corrections](/corrections/) - [The ranking](/ranking/) - [FAQ](/faq/) --- # Privacy policy URL: https://aiagentsecuritysolutions.com/privacy/ Updated: 2026-11-16 The site is static. It sets no cookies and runs no analytics scripts. CloudFront access logs record request details such as IP address, time, page and user agent, and are kept for 90 days. Email sent to the editor is kept only as long as needed to answer or make a correction. ## What data does the site collect? Only server access logs from CloudFront: IP address, request time, page requested, status code, referrer and user agent. They are kept for 90 days and then deleted. They are used to keep the site running and to spot abuse. ## Does the site use cookies or analytics? No. The site sets no cookies and loads no analytics scripts, advertising tags or third-party trackers. The re-rank sliders and the layer map run in your browser and send nothing back. ## What about email? If you write to the editor, the message and your address are used to reply and to check a correction. A correction that is published may cite the source you sent, not your name, unless you ask otherwise. ## How do I ask about my data? Write to [editor@aiagentsecuritysolutions.com](mailto:editor@aiagentsecuritysolutions.com) to ask what is held about a request or to ask for deletion of an email thread. ## Related pages - [Method](/method/) - [Editorial policy](/editorial-policy/) - [Corrections](/corrections/) - [The ranking](/ranking/) - [FAQ](/faq/) --- # Terms of use URL: https://aiagentsecuritysolutions.com/terms/ Updated: 2026-11-16 Datasets and original assets are published under CC BY 4.0: reuse them with credit to AI Agent Security Solutions, edition 16 Nov 2026, and the page URL. Every score and map cell describes public evidence read on that date, and each one links to the source a reader can open and check. ## What can I reuse? The Layer Coverage Map, the Agent Control Point Map and other downloadable datasets may be copied and adapted under CC BY 4.0 with credit. Quote the cite line on the datasets page. Page text may be quoted briefly with a link to the page. ## How current is it? Scores reflect public documentation on the read date, 16 Nov 2026. Agent security products change quickly, so each fact keeps its source link and read date, and a later edition records what changed in the method changelog. ## How are errors corrected? Send the page, the statement and a public source through the corrections page. Confirmed errors are fixed and logged with the date of the fix. ## Related pages - [Method](/method/) - [Editorial policy](/editorial-policy/) - [Corrections](/corrections/) - [The ranking](/ranking/) - [FAQ](/faq/)