Teo BouancheauMCP Servers
YouTube Knowledge MCP
Model Context Protocol server that gives AI assistants the ability to search, analyze, and extract knowledge from YouTube videos. Transcripts, chapters, comments, channel info, and a local library, all over MCP.
← All Projects2026
- Technologies
- Node.js, TypeScript, yt-dlp, Model Context Protocol, Docker
MCP Servers2026
Features
- Search videos by keyword or phrase
- Fetch videos from playlists or channels
- Get video info (title, channel, duration, views, likes, comments, description, tags)
- Extract transcripts (auto-generated or manual captions)
- Get chapters (timestamps and structure)
- Get comments (top comments sorted by popularity)
- Get channel info (name, handle, subscriber count, description)
- List formats (available download resolutions and codecs)
- Download videos with quality selection (local mode only)
- Save to library (summaries, notes, skills) (local mode only)
- List library with tag filtering (local mode only)
Compatibility
Works with Claude Desktop, Claude Code, Claude.ai, Cursor, and any MCP-compatible client. Supports both local (stdio) and remote (Streamable HTTP) transports.
Stack
- Node.js 20+
- yt-dlp for video and metadata extraction
- Model Context Protocol SDK
- Docker support for self-hosted remote mode
Services
Frequently Asked Questions
- What can YouTube Knowledge MCP do?
- It lets AI assistants search YouTube, fetch video and channel info, extract transcripts and chapters, pull comments, and in local mode, download videos and save summaries to a personal library.
- Does it require a cloud service?
- No. It supports local (stdio) mode with self-hosted Docker deployment, as well as remote (Streamable HTTP) mode.
