YouTube Transcript MCP Server

A Model Context Protocol (MCP) server that provides tools and resources for fetching YouTube video transcripts. This server can be deployed to Smithery or run locally.

Features

Tools

• get_transcript: Fetch transcript from YouTube video URL with language and timestamp options
• extract_video_id_tool: Extract video ID from various YouTube URL formats

Resources

• youtube://transcript/{video_id}: Get formatted transcript text by video ID
• youtube://metadata/{video_id}: Get transcript metadata (segments, languages, etc.)

Prompts

• analyze_transcript: Structured prompt for analyzing video content
• summarize_transcript: Configurable prompt for summarizing videos (short/medium/long)

Local Development

1. Install dependencies:

   uv sync
   

2. Test the basic transcript fetcher:

   uv run python main.py "https://www.youtube.com/watch?v=LBn0eAxQpb8"
   

3. Run FastAPI server:

   uv run uvicorn api_server:app --reload
   

4. Test MCP server:

   uv run mcp dev mcp_server.py
   

Deployment to Smithery

Prerequisites

• GitHub repository with this code
• Smithery account

Deploy Steps

1. Push to GitHub:

   git add -A
   git commit -m "Ready for Smithery deployment"
   git push origin main
   

2. Connect to Smithery:

   • Go to Smithery
   • Connect your GitHub repository
   • Navigate to the Deployments tab

3. Configure deployment:

   • The smithery.yaml file is already configured
   • Set any environment variables if needed
   • Click “Deploy”

Configuration Options

The server supports these configuration parameters:

• rateLimit: API requests per minute (1-1000, default: 60)
• languages: Default transcript languages (default: “en”)
• includeTimestamps: Include timing info by default (default: true)
• logLevel: Logging verbosity (DEBUG/INFO/WARNING/ERROR, default: INFO)

API Usage Examples

Using Tools

# Get transcript
result = await session.call_tool("get_transcript", {
    "url": "https://www.youtube.com/watch?v=LBn0eAxQpb8",
    "languages": "en,es",
    "include_timestamps": True
})

# Extract video ID
video_info = await session.call_tool("extract_video_id_tool", {
    "url": "https://youtu.be/LBn0eAxQpb8"
})

Using Resources

# Get transcript content
transcript = await session.read_resource("youtube://transcript/LBn0eAxQpb8")

# Get metadata
metadata = await session.read_resource("youtube://metadata/LBn0eAxQpb8")

Using Prompts

# Analysis prompt
analysis_prompt = await session.get_prompt("analyze_transcript", {
    "video_url": "https://www.youtube.com/watch?v=LBn0eAxQpb8"
})

# Summary prompt
summary_prompt = await session.get_prompt("summarize_transcript", {
    "video_url": "https://www.youtube.com/watch?v=LBn0eAxQpb8",
    "length": "medium"
})

Architecture

graph TD
    A[YouTube URL] --> B[Video ID Extraction]
    B --> C[YouTube Transcript API]
    C --> D[Transcript Processing]
    D --> E[MCP Tools/Resources]
    E --> F[Client Applications]
    
    G[Smithery] --> H[Docker Container]
    H --> I[FastMCP Server]
    I --> J[Streamable HTTP]
    J --> K[MCP Protocol]

Error Handling

The server includes comprehensive error handling for:

• Invalid YouTube URLs
• Videos without transcripts
• Network connectivity issues
• Rate limiting
• Malformed requests

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test locally with uv run mcp dev mcp_server.py
5. Submit a pull request

License

MIT License - see LICENSE file for details.

Support

• MCP Documentation
• Smithery Documentation
• YouTube Transcript API