MCP YouTube Transcript Server

MCP YouTube Transcript Server is an MCP server that retrieves transcripts and metadata directly from YouTube videos for LLM analysis. It connects AI assistants like Claude Desktop, Cursor, and Cline to YouTube video content without requiring external API keys. Content creators, researchers, developers, and analysts use this server to read, summarize, and process video audio text programmatically. The server fetches raw subtitles or auto-generated transcripts using video URLs or IDs. It handles multiple languages, provides automatic paragraph segmentation, and normalizes text by resolving HTML entities and extra whitespace. In addition to speech transcripts, it returns metadata including video title, duration, transcript count, and character counts. This setup enables AI agents to reference specific sections of long videos, compare discussions across talks, or draft written summaries directly from multimedia links.

Category: Design, Media & Creative

Tags: content-analysis, transcripts, video, youtube

Visit MCP YouTube Transcript Server

How to install and configure MCP YouTube Transcript Server

1. Ensure Node.js version 18 or higher is installed on your machine. 2. Open your client configuration file. For Claude Desktop, locate ~/Library/Application Support/Claude/claude_desktop_config.json on macOS or %APPDATA%\Claude\claude_desktop_config.json on Windows. 3. Add the server under the mcpServers object: json { "mcpServers": { "youtube-transcript": { "command": "npx", "args": [ "-y", "@sinco-lab/mcp-youtube-transcript" ] } } } 4. Restart Claude Desktop or your MCP client to apply the changes and start fetching transcripts.

What you can do with MCP YouTube Transcript Server

• Extracting complete video transcripts and structured metadata to generate structured summaries and action items in Claude Desktop. * Translating and reviewing international YouTube videos by requesting transcripts in specific supported language codes. * Formatting video speech into readable paragraphs using paragraph mode for easier reading, documentation, and reference. * Analyzing long lectures, interviews, and webinars to isolate specific topics and discussion points without watching full recordings.

Key facts

• https://github.com/lnmunhoz/mcp-youtube-transcript
• Design, Media & Creative, Web Search & Research
• content-analysis, transcripts, video, youtube

Part of MCP Servers

How do I install MCP YouTube Transcript Server?

You can install it by configuring your MCP client configuration file, such as claude_desktop_config.json, to execute npx -y @sinco-lab/mcp-youtube-transcript under the mcpServers key. Ensure Node.js 18 or higher is installed before starting your client.

What can MCP YouTube Transcript Server do?

The server retrieves text transcripts and video metadata from YouTube videos using their URL or ID. It provides options to select different subtitle languages, formats output as continuous text or segmented paragraphs, cleans up HTML entities, and tracks duration and character count.

Which MCP clients work with MCP YouTube Transcript Server?

The server is compatible with any client supporting the Model Context Protocol standard through stdio, including Claude Desktop, Cursor, Cline, and the official MCP Inspector.

Is MCP YouTube Transcript Server open source?

Yes, MCP YouTube Transcript Server is an open source project distributed under the MIT License on GitHub.