youtube

# YouTube Transcript to Structured Notes ## Arguments ``` $ARGUMENTS ``` ## Instructions This workflow fetches a YouTube video transcript and transforms it into a structured, skimmable reference document. Uses a forked context since the transcript processing doesn't need to inform follow-up conversation. ### 1. Parse Video Input - **Full URL**: Use directly (supports youtube.com, youtu.be) - **Video ID**: Construct `https://youtube.com/watch?v={id}` - **If missing**: Ask for the video URL ### 2. Fetch Video Metadata ```bash yt-dlp --print "%(title)s" --print "%(channel)s" --print "%(uploader)s" --print "%(upload_date)s" --print "%(duration)s" --skip-download "VIDEO_URL" ``` Capture: - **Title** — for filename and frontmatter - **Channel/Uploader** — for attribution (use uploader if personal, channel if organizational) - **Upload Date** — format as YYYY-MM-DD - **Duration** — for context ### 3. Download Transcript ```bash mkdir -p tmp yt-dlp --skip-download --write-auto-subs --sub-langs "en" --sub-format "srt" --convert-subs srt -o "tmp/transcript" "VIDEO_URL" ``` Saves to `tmp/transcript.en.srt`. If no English subs, try `en-orig` or check `--list-subs`. ### 4. Determine Output Filename Use **APA-style title capitalization** with creator: **Format:** `Reference/Full Title Here (Creator Name).md` **APA rules:** - Capitalize first word and all major words - Capitalize words of 4+ letters - Lowercase: a, an, the, and, but, or, for, nor, on, at, to, by (unless fi...