receiving-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of code review feedback, which constitutes a potential attack surface.
- Ingestion points: Feedback handled from both 'human partner' and 'External Reviewers' as defined in SKILL.md.
- Boundary markers: Instructions mandate a multi-step 'Response Pattern' (Read, Understand, Verify, Evaluate, Respond, Implement) to create logical barriers against blind obedience.
- Capability inventory: References using grep for searches and gh api for GitHub Pull Request interactions.
- Sanitization: The skill relies on technical verification against the existing codebase as a logic-level filter rather than programmatic sanitization.
- [COMMAND_EXECUTION]: The skill directs the agent to use specific command-line tools to verify technical requirements and respond to reviews.
- Evidence: SKILL.md explicitly mentions using grep codebase for actual usage and using the GitHub CLI via gh api repos/{owner}/{repo}/pulls/{pr}/comments/{id}/replies for PR thread management. These are standard developer tools used for their intended purpose.
Audit Metadata