5hivanand/mcp-youtube-video-summary

YouTube Video Summary

A Model Context Protocol (MCP) server that fetches YouTube video transcripts and generates customizable summaries.

Features

• Video ID Extraction: Extract YouTube video IDs from various URL formats (youtube.com, youtu.be, etc.)
• Transcript Fetching: Fetch video transcripts using the YouTube Transcript API
• Caching: Cache transcripts locally for faster retrieval
• Customizable Prompts: Generate summaries in multiple formats:
  • Quick 30-second summary
  • Detailed 2-3 minute summary
  • Structured notes in markdown format

Installation

Requirements

• Python >= 3.12
• uv package manager (optional, but recommended)

Setup

1. Clone the repository:

git clone <repository-url>
cd mcp-youtube-video-summary

2. Install dependencies:

# Using uv
uv sync

# Or using pip
pip install -r requirements.txt

Usage

Running the MCP Server

uv run main.py

The server will start with standard I/O transport for MCP communication.

Available Tools

• get_youtube_video_transcript(url: str): Fetches the transcript for a given YouTube URL

Available Prompts

• quick_summary_prompt(video_url: str): Generates a quick 30-second summary
• detailed_summary_prompt(video_url: str): Generates a detailed 2-3 minute summary
• structured_notes_prompt(video_url: str): Generates comprehensive structured notes

Integration with AI Clients

Claude Desktop

1. Open ~/Library/Application Support/Claude/claude_desktop_config.json
2. Add the following configuration:

{
  "mcpServers": {
    "Youtube-Video-Summary": {
      "command": "uv",
      "args": [
        "--directory",
        "/Users/Shiva/Developer/mcp/mcp-youtube-video-summary",
        "run",
        "main.py"
      ]
    }
  }
}

3. Restart Claude Desktop

VS Code

1. Install the MCP extension for VS Code
2. Add the server configuration to VS Code settings:

{
  "mcp.servers": {
    "Youtube-Video-Summary": {
      "command": "uv",
      "args": [
        "--directory",
        "/Users/Shiva/Developer/mcp/mcp-youtube-video-summary",
        "run",
        "main.py"
      ]
    }
  }
}

Project Structure

.
├── main.py                          # MCP server entry point
├── prompts.yaml                     # Prompt templates
├── pyproject.toml                   # Project configuration
├── youtube_transcript/
│   ├── __init__.py                 # Package exports
│   ├── video_id.py                 # Video ID extraction
│   ├── transcript.py               # Transcript fetching
│   └── cache.py                    # Local caching
└── README.md                        # This file

Configuration

Edit prompts.yaml to customize the summary prompts for different use cases.

Dependencies

• fastmcp: Model Context Protocol framework
• youtube-transcript-api: YouTube transcript fetching
• PyYAML: YAML configuration parsing

License

MIT

Contributing

Feel free to submit issues and enhancement requests!