The Agent Skills Directory

[COMMAND_EXECUTION]: The skill leverages shell commands to perform its primary functions. It uses Python helper scripts (doctor.py, probe_media.py, check.py) which invoke ffmpeg and ffprobe via subprocess.run. Additionally, SKILL.md provides approved command recipes for the agent to execute using the Bash tool.
[SAFE]: The skill implements robust safety boundaries, such as mandatory metadata probing before any file mutation and a strict policy against overwriting source files in-place.
[SAFE]: External references are limited to standard system utilities (ffmpeg, ffprobe) and instructions for package managers like brew and uv. No suspicious network activity, data exfiltration, or obfuscation was detected.
[SAFE]: The check.py script references a developer toolkit located in the parent directory (skill-creator), which is consistent with standard development practices for this author's ecosystem.

ffmpeg