get-pr-comments
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes review comments from pull requests, which are an untrusted external source. An attacker could embed instructions in a PR comment to influence the agent's summary or behavior.
- Ingestion points: Pull request review and discussion comments (SKILL.md).
- Boundary markers: Not specified.
- Capability inventory: The skill description only mentions fetching and summarizing data; no dangerous tool usage or scripts are defined in the provided file.
- Sanitization: No explicit sanitization of external content is mentioned.
Audit Metadata