Summarize YouTube videos, store with vector embeddings, and search semantically. Use when the user shares a YouTube link, asks to summarize a video, search their saved video notes, or look up what videos they have about a topic.
These tools are available via the youtube-tools MCP server:
Summarize a YouTube video by URL or video ID. Extracts transcript via yt-dlp, analyzes with Claude, stores in SQLite with vector embeddings.
youtube_summarize({ video: "https://youtube.com/watch?v=dQw4w9WgXcQ" })
youtube_summarize({ video: "dQw4w9WgXcQ" })
Semantic vector search across all stored video summaries. Finds relevant themes, topics, insights, and quotes.
youtube_search({ query: "quantum physics experiments", limit: 5 })
List stored videos with optional filters.
youtube_list({})
youtube_list({ keyword: "physics", channel: "Veritasium", limit: 10 })
Get full details for a specific video.
youtube_get({ video_id: "dQw4w9WgXcQ" })