cleancoding
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed to improve engineering discipline and explicitly prioritizes security and data integrity over aesthetic coding choices.
- [SAFE]: Includes specific instructions for trust boundary validation, advising the agent to treat external content (web, files, tool output) strictly as data rather than instructions, which mitigates indirect prompt injection risks.
- [SAFE]: Mandates the sanitation of logs to prevent credential or secret leakage during debugging and error handling.
- [SAFE]: Implements a 'loop limit' to prevent the agent from performing excessive or repetitive actions, serving as a safeguard against infinite loops or resource exhaustion.
Audit Metadata