# Install via pip
pip install yt-dlp
# Or via homebrew (macOS)
brew install yt-dlp
# Verify installation
yt-dlp --version
# Create temporary directory for this analysis
ANALYSIS_DIR=$(mktemp -d)
echo "Using temporary directory: $ANALYSIS_DIR"
Download transcript using yt-dlp to extract subtitles/transcripts to the temporary directory:
# Download transcript only (no video)
yt-dlp --skip-download --write-auto-sub --sub-format vtt --output "$ANALYSIS_DIR/transcript.%(ext)s" URL
# Or get manually created subtitles if available (higher quality)
yt-dlp --skip-download --write-sub --sub-lang en --sub-format vtt --output "$ANALYSIS_DIR/transcript.%(ext)s" URL
# Get video metadata for context
yt-dlp --skip-download --print-json URL > "$ANALYSIS_DIR/metadata.json"
Handle long transcripts - For transcripts exceeding 8,000 tokens (roughly 6,000 words or 45+ minutes):
- Split into logical chunks based on timestamp or topic breaks
- Generate a summary for each chunk focusing on key concepts
- Create an overall synthesis connecting themes to the user's question
- Reference specific timestamps for detailed sections
For shorter transcripts, analyze directly without chunking.
Provide analysis in this format:
Video Overview:
- Title, author, duration
- Main topic and key themes
Key Insights:
- Concept 1 with timestamp
- Concept 2 with timestamp
- Technical approaches explained
Relevance to Your Project:
- Direct applications
- Differences from current approach
- Potential improvements or learnings
Specific Recommendations:
- Actionable items based on video content
- Code patterns or techniques to consider
Multiple languages: Prefer English transcripts using --sub-lang en. If unavailable, check available languages with --list-subs.
Long processing time: Set expectations for videos over 2 hours. Offer to focus on specific sections if timestamps provided.
Extract concrete examples and code patterns when available. Reference specific timestamps so users can jump to relevant sections.
When comparing with project code, be specific about similarities and differences. Vague comparisons like "similar approach" don't add value.
For technical content, identify the underlying patterns and principles rather than surface-level implementation details. Help users understand transferable concepts.