| name | youtube-text-fetch |
| description | Fetch transcript/caption text from YouTube videos using youtube-transcript-api. Use when: (1) User wants text from a YouTube video, (2) User provides YouTube URLs and needs transcripts, (3) User says 'fetch youtube text', 'get captions', 'youtube transcript'. Outputs plain text saved to the repo-scoped cclogs dir (resolved via get-logdir.js) as youtube-{VIDEO_ID}.txt. |
| user-invocable | true |
| argument-hint | YouTube URL(s) — space-separated if multiple. Optional flags: --timestamps (include timestamps in output) |
YouTube Text Fetch
Fetch transcript/caption text from YouTube videos.
Note: For visual analysis of a YouTube video (extracting frames and reading them
chronologically with vision to produce a timestamped narrative, e.g. debugging a
recording or answering "what happens in this video"), use the /video-reader skill
instead. This skill is for quick transcript-only extraction.
Prerequisites
Requires youtube-transcript-api Python package:
pip3 install youtube-transcript-api
If not installed, install it automatically before proceeding.
Workflow
1. Parse Video IDs
Extract video IDs from provided URLs. Supported formats:
https://youtu.be/<ID>https://youtu.be/<ID>?si=...https://www.youtube.com/watch?v=<ID>
