The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run and subprocess.check_output to execute ffmpeg and ffprobe for video analysis (crop detection) and subtitle burning. These commands are invoked using argument lists, which is a standard security practice to mitigate shell injection risks.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it processes content from external files that could be influenced by an attacker.
Ingestion points: Reads and processes data from timing.json, script.json, and subs_pages.json across split.py and build.py.
Boundary markers: The instructions do not define specific delimiters or "ignore" directives to prevent the agent from interpreting instructions embedded within the subtitle text data.
Capability inventory: The skill maintains capabilities for file system writes and local command execution via subprocess for media processing.
Sanitization: Path inputs are handled through pathlib, and burn.py includes logic in _escape_subs_path to escape characters for the ffmpeg libass filtergraph.

burn-subtitles-cn