The Agent Skills Directory

[SAFE]: No malicious patterns, data exfiltration, or prompt injection attempts were detected in the skill instructions. The content is focused solely on performance profiling, benchmarking, and optimization methodologies.
[COMMAND_EXECUTION]: The skill mentions common, benign CLI tools for performance monitoring (e.g., cProfile, py-spy, perf, flamegraphs) as part of a legitimate developer workflow. No dangerous or hidden command executions are present.
[DATA_EXFILTRATION]: There are no network requests, secret harvesting, or attempts to access sensitive files. The skill operates on the provided code or profiling data within the agent's environment.

performance-expert