review-hog-perspective-contracts-security
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of pull request code chunks. This creates an attack surface where malicious instructions embedded in the code (e.g., in comments or string literals) could potentially influence the agent's behavior during the review process.
- Ingestion points: Pull request chunks and code files processed by the reviewer agent.
- Capability inventory: The skill utilizes the
rg(ripgrep) tool to search and read local files. - Boundary markers: None explicitly defined in this perspective file to isolate the content of the code under review from the agent's instructions.
- Sanitization: No specific sanitization, escaping, or filtering of the code input is mentioned.
Audit Metadata