The Agent Skills Directory

[COMMAND_EXECUTION]: The workflow requires the agent to execute a shell command node tools/scripts/analyze-video.js that incorporates user-provided placeholders <url-or-path> and <project-name>. Without strict validation or sanitization by the agent platform, these inputs could be manipulated to execute arbitrary commands.
[EXTERNAL_DOWNLOADS]: The skill relies on external binaries yt-dlp and ffmpeg to download and process media from external sources, introducing untrusted data into the local environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially malicious content from external video transcripts and metadata, which could be used to influence the agent's behavior during the analysis phase.
Ingestion points: The url-or-path argument supplied by the user pointing to external video resources.
Boundary markers: None identified in the skill instructions to separate or ignore instructions embedded in the external content.
Capability inventory: Shell command execution (node), file writes and directory creation within the video-analysis/ path.
Sanitization: The skill does not describe any sanitization or filtering logic for transcripts or metadata before they are used to generate the final documentation.

analyzing-video