review-changes
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's scripts (scripts/pr.py, scripts/pack.py, etc.) use only Python standard library modules, avoiding supply chain risks from external packages.
- [SAFE]: All shell commands are executed using subprocess calls with list-based arguments, which prevents command injection by avoiding shell parsing.
- [SAFE]: The 'pack.py' and 'snapshot.py' scripts implement directory boundary checks to prevent path traversal when reading or writing files.
- [SAFE]: Instructions in 'SKILL.md' explicitly direct the agent to treat PR descriptions and comments as data rather than instructions, mitigating indirect prompt injection attacks.
- [SAFE]: Network access is performed through authenticated official CLI tools (gh and glab) to communicate with GitHub and GitLab, which is required for its primary functionality.
- [SAFE]: No evidence of credential theft, obfuscation, or unauthorized privilege escalation was found in the analyzed files.
Audit Metadata