The Agent Skills Directory

[SAFE]: The skill provides a rigorous four-phase framework for systematic debugging that emphasizes root cause investigation over symptom-based fixes, which helps maintain system integrity.
[SAFE]: The included utility script find-polluter.sh is a benign tool for identifying state-polluting tests through bisection; it operates locally and does not perform unauthorized network operations.
[SAFE]: Documentation such as defense-in-depth.md explicitly promotes security best practices by instructing the agent to validate data at every architectural layer.
[SAFE]: No hardcoded credentials, API keys, or secrets were found; examples provided in documentation use environment variables and placeholder references appropriately.
[SAFE]: The skill's approach to ingesting external data (logs, error messages, and stack traces) is analytical and structured, reducing the likelihood of the agent being misdirected by content within that data.

systematic-debugging