The Agent Skills Directory

[SAFE]: The skill is instructional, providing a systematic approach to performance tuning without including executable malicious content.
[COMMAND_EXECUTION]: Mentions common diagnostic tools such as EXPLAIN ANALYZE, Lighthouse, and py-spy as part of a measurement methodology; no unsafe or automated execution is defined.
[PROMPT_INJECTION]: No patterns of instruction override, safety filter bypass, or system prompt extraction were detected.
[DATA_EXFILTRATION]: No mechanisms for accessing sensitive files or exfiltrating data to external servers are present.

performance-tuning