MCP YouTube Transcript Server

You can retrieve YouTube transcripts through a Model Context Protocol (MCP) server, enabling multi-language access, paragraph formatting, and metadata for content analysis. This server is designed to be integrated with MCP clients so you can fetch transcripts and related video data efficiently.

How to use

Use a compatible MCP client to connect to the YouTube Transcript MCP server. You will request transcripts by providing a YouTube video URL or ID, and you can optionally specify a language code and whether to enable paragraph mode. The server returns the transcript content along with metadata such as the video title, language, and transcript counts. You can combine this with your analysis workflows to extract insights, generate summaries, or index transcripts for search.

How to install

Prerequisites: Node.js 18 or higher.

{
  "mcpServers": {
    "youtube_transcript": {
      "command": "npx",
      "args": [
        "-y",
        "@sinco-lab/mcp-youtube-transcript"
      ]
    }
  }
}

Quick setup script for macOS:

# Create directory if it doesn't exist
mkdir -p ~/Library/Application\ Support/Claude

# Create or update config file
cat > ~/Library/Application\ Support/Claude/claude_desktop_config.json << 'EOL'
{
  "mcpServers": {
    "youtube-transcript": {
      "command": "npx",
      "args": [
        "-y",
        "@sinco-lab/mcp-youtube-transcript"
      ]
    }
  }
}
EOL

Additional sections

Troubleshooting and maintenance guidelines are provided to help you monitor logs and manage the MCP server environment.

Testing and development notes include how to verify basic operations with Claude or via the MCP Inspector. You can run local checks, list available tools, and test transcript retrieval against a sample YouTube video URL.

Logging and cache maintenance steps cover how to view Claude logs, identify MCP server logs, and clear the npx cache if necessary.

API Reference

get_transcripts fetches transcripts for a given YouTube video. You can specify a language and enable paragraph mode to adjust the text output.

Development

Project structure and core components are organized to handle transcript fetching, text processing, and utilities for dealing with HTML entities, punctuation, and paragraph segmentation.

Contributing

Contributions are welcome. You can submit issues and pull requests to improve features, support more languages, or enhance error handling.

License

This project is released under the MIT License.

Available tools

get_transcripts

Fetches transcripts for a YouTube video, with optional language and paragraph mode, returning transcript content and metadata.