The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run to call ffmpeg and ffprobe for video processing operations. All such calls are implemented using argument lists rather than shell strings, which prevents shell injection vulnerabilities.- [PROMPT_INJECTION]: The skill possesses a minimal attack surface for indirect prompt injection as it processes media files and takes path inputs. However, it does not process natural language instructions from these inputs, and it employs strict validation.
Ingestion points: Path and numeric arguments provided via CLI (e.g., --input, --segment, --audio) in the scripts/ directory.
Boundary markers: Not applicable for binary media transformation tools.
Capability inventory: Executes ffmpeg/ffprobe and writes to the .mediaskills/generated/ directory.
Sanitization: Inputs are resolved as absolute paths, numeric parameters are cast to floats, and the demuxer list file in scripts/concat.py is properly escaped.

video-transformation