The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes ffmpeg and ffprobe to process video and audio files. These operations are performed via Node.js spawnSync using argument arrays, which prevents shell injection. This behavior is appropriate for the skill's primary function.
[INDIRECT_PROMPT_INJECTION]: The skill ingests transcript data from JSON files, which could theoretically contain malicious instructions. However, the scripts process this as data for identifying cuts, and the final decision-making is review-gated by the user.
Ingestion points: scripts/find-cut-candidates.mjs and scripts/apply-cuts.mjs read user-provided transcript files.
Boundary markers: No explicit instruction delimiters or safety warnings are present in the scripts to isolate transcript text from the agent's logic.
Capability inventory: The skill can read/write files and execute system commands for video processing.
Sanitization: The skill performs structural validation of the JSON input but does not sanitize the text content for potential injection strings.

cut-mistakes