tool-systematic-debugging
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- COMMAND_EXECUTION (SAFE): The skill includes a bash utility script ('find-polluter.sh') that automates the execution of local tests via 'npm test' to identify state-polluting files. This is a standard developer tool for diagnostic purposes.
- DATA_EXPOSURE (SAFE): The framework suggests inspecting environment variables and stack traces to identify root causes. These operations are conducted within the local context and do not involve unauthorized data access or exfiltration.
- PROMPT_INJECTION (SAFE): The package includes internal validation scenarios ('test-pressure-1.md' through 'test-pressure-3.md') which are used to test the agent's adherence to the debugging process. They are not malicious prompt injections.
- SAFE (SAFE): The 'defense-in-depth.md' module actively promotes security best practices, such as implementing environment guards to prevent dangerous operations during automated testing.
Audit Metadata