debugging-and-error-recovery
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides structured debugging guidelines and utilizes standard development commands (npm test, npm run build, git bisect) which are appropriate for its stated purpose.
- [PROMPT_INJECTION]: The skill demonstrates proactive security measures by including a 'Treating Error Output as Untrusted Data' section. This instructs the agent to analyze error logs for diagnostic purposes only and never to execute instructions or navigate to URLs found within them, effectively addressing the risk of indirect prompt injection via external logs.
Audit Metadata