review-hog-validation-criteria
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes Pull Request (PR) data, which is an untrusted external source that could contain malicious instructions.
- Ingestion points: The agent is instructed to read "flagged code" and "live codebase" associated with PRs, which includes content written by potentially untrusted contributors (SKILL.md).
- Boundary markers: The instructions do not define specific delimiters or "ignore instructions" markers to help the agent distinguish between its logic and the code it is analyzing.
- Capability inventory: The skill instructions imply the agent has the capability to read files and traverse a codebase to verify reachability and call sites.
- Sanitization: No sanitization, filtering, or escaping mechanisms are described for the external code or comments being processed.
Audit Metadata