Microsoft AI agent security
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.
Facts read on .
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.
| 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, microsoft-14, microsoft-19, microsoft-54, 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, 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, microsoft-30, microsoft-31, 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, microsoft-34, microsoft-35, 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, 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, microsoft-30, microsoft-34, 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, microsoft-31, microsoft-36, 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, microsoft-35, microsoft-38, 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, microsoft-34, microsoft-36, microsoft-37, 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, microsoft-52, 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, microsoft-64, microsoft-57 |
Where it ranks
| View | Rank | Total |
|---|---|---|
| Overall ranking | 3rd | 63.0 |
| Ranked for the endpoint layer | 2nd | 69.0 |
| Ranked for the browser layer | 9th | 51.0 |
| Ranked for the cloud workspace layer | 4th | 66.0 |
| Ranked for the identity layer | 2nd | 66.0 |
| Ranked for the gateway layer | 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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .
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, read .