agent-introspection-debugging
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and defines a workflow for error handling and self-debugging. It does not contain executable code, external dependencies, or commands that interact with the system or network. It encourages best practices such as least privilege, human escalation for high-risk failures, and evidence-based recovery.
Audit Metadata