audit-fix-loop
Pass
Audited by Gen Agent Trust Hub on Apr 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a structured workflow for quality audits without malicious intent.
- [SAFE]: Command execution is restricted to local directory initialization (mkdir) for state management.
- [SAFE]: Potential for indirect prompt injection exists as the skill processes untrusted user artifacts, but this is inherent to its primary function and managed within subagent contexts.
Audit Metadata