debugging-and-error-recovery

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured debugging instructions and utilizes standard development tools like 'npm' and 'git' in a manner consistent with its stated purpose.
  • [PROMPT_INJECTION]: The skill includes a 'Treating Error Output as Untrusted Data' section that correctly identifies the risk of malicious instructions being embedded in diagnostic outputs and explicitly instructs the agent to verify such content before execution, acting as a defensive measure.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 12:49 PM
Security Audit — agent-trust-hub — debugging-and-error-recovery