review-methodology
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary purpose is the review and auditing of external, untrusted code and documentation, which creates an attack surface for indirect prompt injection.
- Ingestion points: The skill guides agents to ingest external code diffs, pull requests, and architectural documents for analysis (referenced in SKILL.md and multiple reference files).
- Boundary markers: The provided reporting templates do not include explicit instructions or markers to prevent the agent from following instructions embedded within the code under review.
- Capability inventory: The skill consists purely of markdown files and templates; it contains no executable code, scripts, or direct tool invocations.
- Sanitization: No sanitization logic or filtering is implemented for the data being ingested.
- [SAFE]: All dangerous code patterns mentioned in the references (such as hardcoded credentials or shell injection examples) are used purely for instructional purposes to train the agent on what to detect during audits and do not represent active threats within the skill itself.
Audit Metadata