YouTube Video Summarizer MCP Server

April 1, 2025 ยท View on GitHub

An MCP (Model Context Protocol) server that summarizes YouTube videos by extracting and analyzing their transcripts.

Features

  • Extract metadata and transcripts from YouTube videos
  • Summarize video content using various algorithms:
    • Basic: Simple text extraction and frequency analysis
    • Advanced: TF-IDF based algorithm that creates more coherent summaries (no API needed)
    • OpenAI: Uses OpenAI's API for high-quality summaries (requires API key)
  • Extract key points from the video
  • Provide formatted summaries that can be used by AI assistants
  • Robust error handling with detailed error messages
  • Fallback mechanisms for handling restricted or problematic videos
  • Comprehensive logging for easier troubleshooting

Prerequisites

  • Node.js (v16 or higher)
  • npm or yarn
  • TypeScript

Installation

  1. Clone the repository or download the source code
  2. Navigate to the project directory
  3. Install dependencies:
npm install
  1. Build the project:
npm run build

Configuration

Create a .env file in the root of the project with the following variables:

# YouTube API Key (optional - for additional metadata)
YOUTUBE_API_KEY=

# OpenAI API Key (optional - for AI-based summarization)
OPENAI_API_KEY=

# Maximum length of video to process (in minutes)
MAX_VIDEO_LENGTH=60

Note: The OpenAI API key is optional. Without it, the system will use the advanced TF-IDF based summarizer which still provides good results without requiring any external API.

Troubleshooting

If you're having issues with the YouTube video summarizer, check the TROUBLESHOOTING.md file for common problems and solutions.

Common Issues with YouTube Videos

Some videos may not work with this summarizer due to various restrictions:

  1. Age-restricted videos: Videos that require age verification cannot be accessed without authentication
  2. Private videos: Videos marked as private are not accessible
  3. Region-blocked videos: Videos that are restricted in certain regions may not be accessible
  4. Missing transcripts: Some videos don't have captions or transcripts available
  5. Copyright-restricted videos: Some videos have special copyright protections

The tool now includes improved error handling to help identify these issues more clearly and provide better feedback.

Usage

As a standalone tool

You can run the tool directly from the command line:

node build/index.js

Integrating with Claude or other AI assistants

Configure your Claude Desktop MCP server configuration file to include this MCP server:

{
  "mcpServers": {
    "youtube-video-summarize": {
      "command": "node",
      "args": [
        "/ABSOLUTE/PATH/TO/youtube-video-summarize/build/index.js"
      ],
      "env": {
        "MAX_VIDEO_LENGTH": "60"
      }
    }
  }
}

Available Tools

The MCP server provides the following tools:

summarize-video

Summarizes a YouTube video by extracting its transcript and generating a concise summary with key points.

Parameters:

  • videoUrl: URL of the YouTube video to summarize
  • summarizerType: Type of summarizer to use (basic, advanced, or openai)

get-video-info

Retrieves basic information about a YouTube video without generating a summary.

Parameters:

  • videoUrl: URL of the YouTube video

get-video-transcript

Retrieves the full transcript of a YouTube video.

Parameters:

  • videoUrl: URL of the YouTube video

Summarization Methods

Basic Summarizer

The Basic Summarizer uses word frequency and position-based heuristics to extract important sentences from the transcript. It's fast and requires no external APIs.

Advanced Summarizer

The Advanced Summarizer uses TF-IDF (Term Frequency-Inverse Document Frequency) to identify important sentences. It also employs several techniques to ensure coherence and minimize redundancy. This summarizer provides good results without requiring any external API.

OpenAI Summarizer

If you provide an OpenAI API key, the system can use OpenAI's powerful language models to generate higher-quality summaries. This option requires an API key and internet connectivity.

Architecture

This project follows SOLID principles:

  • Single Responsibility Principle: Each class has a single responsibility
  • Open/Closed Principle: Code is open for extension but closed for modification
  • Liskov Substitution Principle: Different summarizer implementations can be used interchangeably
  • Interface Segregation Principle: Interfaces are specific to their clients' needs
  • Dependency Inversion Principle: High-level modules depend on abstractions

License

MIT