The Agent Skills Directory

[SAFE]: The skill performs legitimate video processing tasks and does not contain any malicious patterns such as prompt injection, data exfiltration, or obfuscation.\n- [SAFE]: Subprocess calls to ffmpeg and ffprobe in scripts/extract_frames.py are implemented using lists rather than shell strings, which is a secure practice to prevent command injection.\n- [SAFE]: The script manages files securely by using the standard tempfile module for temporary output and does not attempt to access sensitive system files or environment variables.\n- [SAFE]: No external Python or Node.js dependencies are required, and the skill does not perform any network requests or remote code downloads.

video-frames