pr-review-panel
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content from GitHub pull requests and issues, creating a surface for indirect prompt injection.\n
- Ingestion points: External data is ingested in SKILL.md via
gh pr diff,gh pr view --comments, andgh issue viewcommands.\n - Boundary markers: There are no specific delimiters or instructions to treat the diff or comment text as data only, increasing the risk that embedded instructions could be followed.\n
- Capability inventory: The skill has the capability to write back to GitHub using
gh pr reviewand usescodex execfor model inference.\n - Sanitization: No explicit sanitization or filtering of the external content is mentioned.\n
- Mitigation: The skill enforces a mandatory human-in-the-loop gate, requiring explicit approval before posting any results to GitHub, which significantly mitigates the impact of a potential injection.
Audit Metadata