receiving-code-review
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- Indirect Prompt Injection (INFO): The skill's primary function is to process external code review feedback, which represents a potential indirect prompt injection surface. The skill proactively mitigates this risk by defining a mandatory reasoning framework that involves verifying feedback against 'codebase reality' and forbidding performative agreement. Evidence Chain: (1) Ingestion Point: GitHub pull request comments and reviewer feedback. (2) Boundary markers: Present as explicit verification and understanding steps. (3) Capability inventory: Code implementation and GitHub API access. (4) Sanitization: Verification against existing code and logic checks.
- Data Exposure & Exfiltration (INFO): The skill references standard use of the GitHub API for replying to inline pull request comments. No evidence of sensitive credential exposure, unauthorized file access, or exfiltration of private data was detected.
Audit Metadata