DY

dogfrogfog/youtube-analytics-mcp

Developer tools

このページは AI により自動翻訳されています。技術的な詳細は原文をご確認ください。 原文を見る

5 stars 6 forks 品質 39 トレンド 13

Youtube analytics MCP server

概要

A Model Context Protocol (MCP) server for YouTube Analytics data access with demographics and discovery tools, built with a scalable config-driven architecture. - : Get comprehensive channel overview, growth patterns, and vital signs - : Analyze individual video metrics, audience retention, and drop-off points - : Access age/gender breakdowns and geographic distribution data - : Understand traffic sources and search terms driving views - : Track likes, comments, shares, and viewer interaction patterns - : Identify exact moments where viewers drop off for content optimization - : Compare metrics between different time periods - : Research competitor channels and trending content This MCP server follows a that provides: - : Clear separation between tool definitions and implementation - : Easy to add new tools without modifying core server logic - : Standardized error handling and response formatting - : Clean, declarative configuration that serves as documentation To use this...

README

以下は原文 README の抜粋です。原文を見る。

YouTube Analytics MCP Server

A Model Context Protocol (MCP) server for YouTube Analytics data access with demographics and discovery tools, built with a scalable config-driven architecture.

Core Features

  • Channel Analytics: Get comprehensive channel overview, growth patterns, and vital signs
  • Video Performance: Analyze individual video metrics, audience retention, and drop-off points
  • Viewer Demographics: Access age/gender breakdowns and geographic distribution data
  • Discovery Insights: Understand traffic sources and search terms driving views
  • Engagement Metrics: Track likes, comments, shares, and viewer interaction patterns
  • Audience Retention: Identify exact moments where viewers drop off for content optimization
  • Performance Comparison: Compare metrics between different time periods
  • Public Channel Analysis: Research competitor channels and trending content

Core Architecture Principles

This MCP server follows a config-driven architecture that provides:

  • Maintainability: Clear separation between tool definitions and implementation
  • Scalability: Easy to add new tools without modifying core server logic
  • Consistency: Standardized error handling and response formatting
  • Readability: Clean, declarative configuration that serves as documentation

Setup

1. Google API Credentials

To use this YouTube Analytics MCP server, you need to set up Google API credentials:

  1. Go to the Google Cloud Console
  2. Create a new project or select an existing one
  3. Enable the YouTube Analytics API and YouTube Data API v3
  4. Go to “Credentials” and create a new OAuth 2.0 Client ID
  5. Download the credentials as JSON
  6. Save the file as credentials.json in the src/auth/ directory

Privacy Note: All data processing happens locally on your computer. Your credentials and analytics data never leave your machine - the server runs entirely locally and connects directly to Google’s APIs from your system.

2. Development

# Install dependencies
npm install

# Build the project
npm run build

# Run in development mode
npm run dev

# Inspect with MCP Inspector
npm run inspect

Architecture Overview

Project Structure

src/
├── index.ts                 # Main server entry point (config-driven)
├── tool-configs.ts         # Central tool configuration aggregator
├── types.ts                # TypeScript interfaces and types
├── auth/
│   ├── tool-configs.ts     # Authentication tool configurations
│   └── ...
├── server/
│   ├── info-configs.ts     # Server info tool configurations
│   └── ...
└── youtube/tools/
    ├── channel-configs.ts  # Channel analysis tool configurations
    ├── health-configs.ts   # Channel health tool configurations
    ├── audience-configs.ts # Audience demographics tool configurations
    ├── discovery-configs.ts # Traffic source tool configurations
GitHub で README 全文を見る

インストール

This server does not publish a one-line install command.

Open the repository installation guide MCP 初心者?mcpskill.org のガイドから始めましょう →