code-review
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
gh(GitHub CLI) andgitcommand-line tools to gather pull request metadata, diffs, and repository state. It also includes a local Python script,scripts/quality_gates.py, which executesgitcommands viasubprocess.runto perform static analysis of exports and configuration liveness. These operations are essential for the skill's primary function and are implemented using safe practices, such as passing command arguments as a list to avoid shell injection vulnerabilities. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from external pull requests and project documentation. This creates a surface for indirect prompt injection if malicious instructions are embedded within the code or comments under review. The skill mitigates this risk through strict 'read-only' instructions, requiring the agent to ignore any editing tools and focus solely on generating feedback for the user to review. It also emphasizes the importance of manual verification and tracing findings back to repository standards.
Audit Metadata