debugging
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Comprehensive analysis of the 14 files revealed no malicious patterns, obfuscation, or security vulnerabilities. The skill adheres to its stated purpose of providing a systematic debugging framework.
- [COMMAND_EXECUTION]: The skill provides examples of using the Bash tool for system diagnostics, such as checking environment variables and code signing identities. These are legitimate activities within the context of technical troubleshooting.
- [PROMPT_INJECTION]: The skill employs instructional reinforcement ('Iron Laws') to ensure the AI follows the debugging process. These instructions do not attempt to override safety protocols or previous system instructions.
- [SAFE]: The skill's data ingestion surface is intended for diagnostic analysis. Although boundary markers and sanitization are absent, the risk is inherent to debugging. Ingestion points: project files via Read tool, command output via Bash tool; Boundary markers: None; Capability inventory: Read, Write, Edit, Bash, Glob, Grep; Sanitization: None.
Audit Metadata