801-regulations-eu-ai-act
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for regulatory compliance and architecture review. It provides instructional guidance and templates rather than performing automated system modifications.
- [SAFE]: Includes explicit safety constraints requiring the redaction of sensitive data, passwords, and API keys, replacing them with
[REDACTED_SECRET]to prevent accidental data exposure. - [SAFE]: References to external documentation point exclusively to the official European Union legislative portal (eur-lex.europa.eu), which is a trusted and well-known government service.
- [SAFE]: The workflow incorporates defensive instructions, such as prioritizing 'trusted local project evidence' and explicitly warning the agent not to treat free-form external inputs as authoritative instructions, which mitigates potential indirect prompt injection risks from reviewed content.
Audit Metadata