verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and contains no executable code, scripts, or binary files. It focuses on enforcing a reliable workflow for software development tasks.
- [SAFE]: No network operations, external downloads, or sensitive file access patterns were detected. The instructions reference standard development tools (VCS diffs, linters, test runners) without providing malicious implementations.
- [SAFE]: No prompt injection or obfuscation techniques were found. The assertive language used in the document is intended to improve the accuracy and honesty of the agent's reporting rather than bypassing safety controls.
- [SAFE]: The skill does not include any hardcoded credentials or data exfiltration vectors.
Audit Metadata