The Agent Skills Directory

[COMMAND_EXECUTION]: The skill documentation instructs the user to run a local installation script ./install.sh and utilizes a rendering engine called render.py. These scripts are used to manage system environments and execute the final video composite.
[EXTERNAL_DOWNLOADS]: The setup process relies on several well-known external tools and services, including ffmpeg, Node.js, Google Chrome, and the ElevenLabs API. These dependencies are standard for the skill's stated purpose of video and audio processing.
[PROMPT_INJECTION]: The skill processes video transcripts to determine edit points, creating a potential surface for indirect prompt injection where malicious instructions could be embedded in the footage.
Ingestion points: The agent reads video transcripts and timeline images during the 'Transcribe + pack' stage.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the transcripts.
Capability inventory: The skill has the capability to write project metadata, generate EDL files, and execute subprocesses for video rendering.
Sanitization: There is no mention of filtering, escaping, or validating the transcript text before it is processed by the LLM.

video-studio