# 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Guides, glossary, news and the FAQ do not quote ranks or scores; the site prints those from the scores files.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/)
