safety-guard
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a safety framework to monitor and intercept high-risk shell commands (e.g.,
rm -rf,sudo,docker prune) and database operations. - [SAFE]: The skill provides directory-level write restrictions to prevent unauthorized file modifications outside of designated project paths.
- [SAFE]: The implementation involves intercepting standard tool calls to validate safety before execution, which is a defensive security measure.
- [SAFE]: No malicious patterns, obfuscation, or external network activities were detected in the instructions.
Audit Metadata