The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute external media processing binaries. Python scripts including batch_resize.py, media_convert.py, and video_optimize.py use subprocess.run to invoke ffmpeg, ffprobe, and magick. Similarly, the shell scripts batch-remove-background.sh and remove-background.sh execute the rmbg command to process images.
[EXTERNAL_DOWNLOADS]: The shell scripts batch-remove-background.sh and remove-background.sh include a convenience feature that checks for the rmbg CLI and attempts to install it globally using npm install -g rmbg-cli from the official NPM registry if it is not found on the system.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided multimedia files, which constitutes a vulnerability surface for indirect prompt injection via malicious file metadata or content.
Ingestion points: Input media files (images, audio, video) processed by scripts in the scripts/ directory.
Boundary markers: None identified in the logic for handling file inputs.
Capability inventory: The skill can read from and write to the filesystem, and execute shell commands through the invoked media tools.
Sanitization: Python scripts use list-based arguments for subprocess.run to prevent shell command injection, and shell scripts employ variable quoting to handle filenames safely.

media-processing