YouTube MCP Server brings AI-readable YouTube data to Claude Desktop
A community-built MCP server by sfiorini lets AI language models retrieve YouTube video details, transcripts, channel info, playlists, and search results via the YouTube Data API.
Score breakdown
The server gives AI models like Claude a standardized, structured path to YouTube's content layer — transcripts, metadata, and search — without requiring custom API integration work from the developer.
- 01Built by sfiorini; listed on Glama as a JavaScript-based remote MCP server
- 02Supports video details, statistics (views, likes, comments), transcripts with timestamps, channel info, playlists, and search
- 03Transcript retrieval supports multiple languages; default language is configurable via the YOUTUBE_TRANSCRIPT_LANG environment variable
The YouTube MCP Server is a JavaScript-based, remote MCP server implementation authored by sfiorini and catalogued on Glama. It exposes YouTube platform data to AI language models through a standardized Model Context Protocol interface, covering video information (title, description, duration, statistics such as views, likes, and comments), transcript retrieval with timestamp support and multi-language options, channel details and playlists, and cross-platform video search. A recent update added `url` and `videoId` fields to video responses for easier downstream integration.
All tools include capability annotations (read-only, idempotent) intended to improve LLM performance.
The server provides direct resource URIs — for example, `youtube://transcript/{videoId}` for transcript access — and two built-in prompts: `summarize-video` for automated video summarization workflows and `analyze-channel` for channel content strategy analysis. All tools include capability annotations (read-only, idempotent) intended to improve LLM performance. Setup requires a `YOUTUBE_API_KEY` environment variable and supports installation via `npm install -g @sfiorini/youtube-mcp`, npx, or Smithery's CLI (`npx -y @smithery/cli@latest install @sfiorini/youtube-mcp --client claude`), with configuration added to the Claude Desktop JSON config file on macOS or Windows. VS Code integration is also supported.
Key facts
- 01Built by sfiorini; listed on Glama as a JavaScript-based remote MCP server
- 02Supports video details, statistics (views, likes, comments), transcripts with timestamps, channel info, playlists, and search
- 03Transcript retrieval supports multiple languages; default language is configurable via the YOUTUBE_TRANSCRIPT_LANG environment variable
- 04Resource URI youtube://transcript/{videoId} provides direct transcript access
- 05Includes two built-in prompts: summarize-video and analyze-channel
- 06Installable via npm, npx, or Smithery CLI; requires a YOUTUBE_API_KEY environment variable
- 07Reports a 90% Smithery quality score
Topics
Summary and scoring are generated automatically from the original article. We always link back to the publisher and never republish images or paywalled content. Last processed Jun 15, 2026 · 11:57 UTC. How this works →