The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/live_clip.py executes system utilities ffmpeg and ffprobe via the subprocess module to calculate video duration, detect silences, and export video segments.
Evidence: subprocess.check_output and subprocess.run are used in the duration, propose, and export_clip functions with list-based arguments to perform media operations.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied SRT transcript files to identify video highlights, creating a surface for potential instruction injection.
Ingestion points: The agent is prompted in SKILL.md to read SRT content pasted into the prompt or provided via file input to generate a cuts.csv file.
Boundary markers: The prompt template in SKILL.md does not include specific delimiters or instructions to ignore potential commands embedded within the transcript text.
Capability inventory: The skill has the ability to execute file system writes and video processing commands via scripts/live_clip.py based on the agent's identification of highlights.
Sanitization: No textual sanitization or validation is performed on the transcript content before it is processed by the agent, allowing embedded instructions to potentially influence the selection and naming of clips.

live-stream-clips