The Agent Skills Directory

[COMMAND_EXECUTION]: The Python scripts use the subprocess module to call ffmpeg and ffprobe for media processing and metadata extraction.
Evidence: subprocess.run is used in scripts/build.py and scripts/song.py to execute command-line utilities. These calls use argument lists rather than shell strings, which is a secure practice to prevent command injection.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-provided configuration files and media assets, which serves as an attack surface for indirect instructions.
Ingestion points: edit.json and audio/video files processed in scripts/build.py and scripts/song.py.
Boundary markers: The scripts do not use explicit boundary markers to delimit data from instructions in the ingested files.
Capability inventory: The skill has the capability to read local files, execute shell commands (ffmpeg, ffprobe), and write output files to the local system.
Sanitization: The scripts use standard libraries like pathlib for path management and json for parsing structured configuration, providing basic validation against malformed inputs.

scenario-seedance-music-video