skill-auditor
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary purpose is security auditing, which justifies the use of file reading and local script execution for analysis.
- [COMMAND_EXECUTION]: The skill uses local shell commands for repository management (git) and to run its internal scanner script. These actions are restricted to a defined allowlist and are essential for functionality.
- [EXTERNAL_DOWNLOADS]: Remote skills are fetched from GitHub for analysis. This process is protected by strict URL validation and isolation in temporary directories to prevent environment contamination.
- [INDIRECT_PROMPT_INJECTION]: The instructions explicitly direct the agent to treat all audited content as data rather than instructions, providing a robust defense against potential injection attacks hidden in target skills.
- [PROMPT_INJECTION]: Static detections of injection patterns in references/security-checklist.md are false positives, as these strings are part of a detection library used by the auditor and are not executable instructions.
Audit Metadata