The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/analyze_context.py script uses os.system() to check for the presence of CLI tools like gh, docker, and python3 during a health check. While os.system() is flagged for execution, it is used here with hardcoded strings (which gh > /dev/null 2>&1) to verify installation, which is a common and safe practice in this context.
[COMMAND_EXECUTION]: The skill provides instructions to execute local Python scripts and shell commands (cp, ln, chmod) for installation and management. These operations are restricted to the local environment and are consistent with the skill's stated purpose of managing other local skills.
[SAFE]: The garbage_collect function in scripts/analyze_context.py performs file deletions (path.unlink()), but it is limited to specific, non-sensitive temporary file patterns (*.skill-temp.*, .claude-export-*.pdf, diagnose-*.log) within the project root.

skills-smart-manager