/watch — YouTube → Knowledge Pipeline

Eyes that see. Ears that listen. Knowledge that stays.

Extract transcripts from YouTube videos using yt-dlp, then optionally pipe through /learn for deep analysis.

Usage

/watch <url>                    # Extract CC + summarize
/watch <url> --raw              # Extract CC only, save raw SRT
/watch <url> --learn            # Extract CC → /learn --deep pipeline
/watch <url> --summary          # Extract CC → concise summary only

---

Step 0: Validate & Detect

date "+🕐 %H:%M %Z (%A %d %B %Y)" && ORACLE_ROOT=$(git rev-parse --show-toplevel 2>/dev/null)
if [ -n "$ORACLE_ROOT" ] && [ -f "$ORACLE_ROOT/CLAUDE.md" ] && { [ -d "$ORACLE_ROOT/ψ" ] || [ -L "$ORACLE_ROOT/ψ" ]; }; then
  PSI="$ORACLE_ROOT/ψ"
else
  ORACLE_ROOT="$(pwd)"
  PSI="$ORACLE_ROOT/ψ"
fi

Check yt-dlp

if command -v yt-dlp &>/dev/null; then
  YTDLP="yt-dlp"
elif [ -x /tmp/yt-dlp ]; then
  YTDLP="/tmp/yt-dlp"
else
  echo "⚠️ yt-dlp not found. Install: pip install yt-dlp OR curl -L https://github.com/yt-dlp/yt-dlp/releases/latest/download/yt-dlp -o /tmp/yt-dlp && chmod +x /tmp/yt-dlp"
  exit 1
fi

Validate URL

Extract video ID from URL. Accept:

• https://www.youtube.com/watch?v=XXXXX
• https://youtu.be/XXXXX
• https://youtube.com/watch?v=XXXXX

VIDEO_URL="$1"
VIDEO_ID=$(echo "$VIDEO_URL" | grep -oP '(?:v=|youtu\.be/)[\w-]{11}' | head -1 | sed 's/v=//')
if [ -z "$VIDEO_ID" ]; then
  echo "❌ Invalid YouTube URL: $VIDEO_URL"
  exit 1
fi
echo "🎬 Video ID: $VIDEO_ID"

---

Step 1: Extract Video Metadata

$YTDLP --print title --print duration_string --print channel --skip-download "$VIDEO_URL" 2>/dev/null

Save as variables: , , .