The Agent Skills Directory

[SAFE]: The skill establishes a systematic framework for debugging and repair ("Nerd Surgery") that prioritizes evidence over assumptions. It includes detailed diagnostic mappings and internal loops to ensure logical and safe operations.\n- [SAFE]: Robust safety instructions are included, such as "Never request a secret in chat", "ask for access setup, never credentials", and requirements for concise approval before performing remote writes or production deployments.\n- [SAFE]: No patterns of prompt injection, data exfiltration, or persistence mechanisms were detected. The instructions are focused entirely on the stated purpose of technical diagnosis and repair.\n- [SAFE]: No obfuscated content, homoglyph attacks, or suspicious external references were identified in the skill instructions or supporting documentation.

nerd-surgery