The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Python's subprocess module in several utility scripts (scripts/run_eval.py, scripts/improve_description.py, eval-viewer/generate_review.py) to execute shell commands such as claude -p for model interactions and lsof for port management. These are internal tools used for the skill's primary development and testing purpose.
[DYNAMIC_EXECUTION]: The skill generates and executes benchmarks and evaluations by spawning subagents. It creates temporary command files in .claude/commands/ to test skill triggering behavior. This is a core part of the skill's functionality for optimizing agent skill descriptions.
[SAFE]: The skill's own code behavior, including network operations for the evaluation viewer and local file manipulation for workspace management, is consistent with its stated purpose as a developer tool. No data exfiltration or malicious patterns were detected.

skill-creator