council
Pass
Audited by Gen Agent Trust Hub on Nov 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data in the form of code review feedback and scores from multiple external 'delegate' models to automate code modifications and update documentation.
- Ingestion points: Feedback, scores, and findings provided by external models (delegates) during the code review process (referenced in SKILL.md).
- Boundary markers: Employs a 'Verify and merge' step to check findings against the code before execution (referenced in SKILL.md).
- Capability inventory: File system write access (for applying code fixes) and pull request description modification.
- Sanitization: Utilizes a verification step to filter or validate claims from reviewers before summarizing findings or applying code changes.
Audit Metadata