code-review
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill establishes a comprehensive defensive framework for code quality. Point 6 specifically mandates checking for hardcoded secrets, SQL injection, and proper input validation.
- [SAFE]: No malicious obfuscation, persistence mechanisms, or unauthorized network operations were detected.
- [SAFE]: The instructions do not trigger remote code execution or package installation; it uses the permitted toolset solely for analyzing existing code repositories.
- [SAFE]: While the skill processes external input (code reviews), the risk of indirect prompt injection is mitigated by its defensive purpose and instructions to use provided context artifacts rather than executing arbitrary code.
Audit Metadata