clean-code-guard

Pass

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and provides a framework for code quality and review based on established engineering principles.- [PROMPT_INJECTION]: Instructions focus on improving code quality and do not contain patterns aimed at bypassing safety filters or overriding core agent behavior.- [EXTERNAL_DOWNLOADS]: References external documentation and research papers from well-known sources including arXiv.org, Martin Fowler's blog, and official GitHub repositories.- [DATA_EXFILTRATION]: No network access is requested in the tool configuration, and the instructions do not suggest exfiltrating data to external servers.- [REMOTE_CODE_EXECUTION]: The skill does not involve script execution, package installation, or remote code downloads. It specifically advises against adding new dependencies for trivial tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 14, 2026, 09:51 PM
Security Audit — agent-trust-hub — clean-code-guard