Google Search MCP Server

Google Search MCP Server

An MCP server that provides Google Search functionality with automatic API key rotation and intelligent quota management, enabling natural language search queries with advanced filtering options.

Category
访问服务器

README

Google Search MCP Server

A Model Context Protocol (MCP) server that provides Google Search functionality with automatic API key rotation and intelligent quota management.

Features

  • Official Google Custom Search API integration
  • Automatic API key rotation for increased daily limits
  • Persistent quota tracking across sessions and directories
  • Multi-language and geolocation support
  • Advanced search filters (date, file type, site-specific)
  • SafeSearch content filtering
  • Global configuration - works from anywhere
  • Compatible with Claude Desktop and other MCP clients

Installation

Quick Setup (Recommended)

# Install the package globally
npm install -g @kyaniiii/google-search-mcp

# Run interactive setup
google-search-setup

This will:

  • ✅ Configure your Google API keys interactively
  • ✅ Set up global configuration file
  • ✅ Automatically integrate with Claude Code
  • ✅ Enable 100+ free searches per day per API key

Advanced Installation

git clone https://github.com/Fabien-desablens/google-search-mcp.git
cd google-search-mcp
npm install
npm run build
npm run setup

Uninstallation

Complete Removal

# Remove the package
npm uninstall -g @kyaniiii/google-search-mcp

# Remove from Claude Code
claude mcp remove google-search

# Remove configuration file (if desired)
npm run clean

Note: Configuration file (~/.google-search-mcp.json) is always preserved during updates and uninstallation to protect your API keys. Use npm run clean only if you want to completely remove your configuration.

Configuration

The setup tool creates a global configuration file at:

  • Windows: %USERPROFILE%\.google-search-mcp.json
  • Linux/macOS: ~/.google-search-mcp.json

This file contains:

  • ✅ Your API keys and search engine IDs
  • Persistent quota tracking (survives restarts)
  • ✅ Automatic daily reset at midnight UTC
  • ✅ Cross-directory usage (works from anywhere)

Example Configuration

{
  "keys": [
    {
      "id": "key_1",
      "apiKey": "AIzaSy...",
      "searchEngineId": "abc123...",
      "dailyUsage": 45,
      "dailyLimit": 100,
      "lastReset": "2024-07-30",
      "isActive": true
    }
  ],
  "lastUpdated": "2024-07-30T15:30:00Z",
  "version": "1.0.0"
}

⚠️ Important: If you manually edit the configuration file (~/.google-search-mcp.json), you must restart Claude Desktop for the changes to take effect.

Getting Google API Credentials

1. Create a Google Cloud Project

  1. Go to Google Cloud Console
  2. Create a new project or select existing
  3. Enable the "Custom Search API"

2. Generate API Key

  1. Navigate to "APIs & Services" > "Credentials"
  2. Click "Create Credentials" > "API Key"
  3. Copy the generated key

3. Create Custom Search Engine

  1. Visit Google Programmable Search Engine
  2. Click "Get started" or "Add"
  3. For "Sites to search", enter * to search the entire web
  4. Copy the Search Engine ID

4. Scale with Multiple Keys

  • Each Google Cloud project provides 100 free searches/day
  • Create multiple projects for more quota (e.g., 3 projects = 300 searches/day)
  • The server automatically rotates between available keys

Usage with Claude Code

After running the setup, the server is automatically configured in Claude Code. You can immediately use it:

User: "Search for latest AI news from this week"
Claude: I'll search for the latest AI news using Google Search...

The server exposes a google_search tool with these parameters:

Required Parameters

  • query (string): Search query

Optional Parameters

  • num (number): Number of results (1-10, default: 5)
  • start (number): Starting index for results
  • safe (string): SafeSearch level ('off', 'active')
  • lr (string): Language restriction (e.g., 'lang_en', 'lang_fr')
  • gl (string): Geolocation (country code: 'us', 'fr', 'uk')
  • dateRestrict (string): Time period ('d1', 'w1', 'm1', 'y1')
  • fileType (string): File type filter ('pdf', 'doc', 'ppt')
  • siteSearch (string): Search specific site
  • siteSearchFilter (string): Include ('i') or exclude ('e') site
  • cr (string): Country restriction ('countryUS', 'countryFR')
  • exactTerms (string): Exact phrase to include
  • excludeTerms (string): Terms to exclude
  • orTerms (string): Alternative terms (OR search)
  • sort (string): Sort by date ('date')
  • searchType (string): Search type ('image' for image search)

Examples

Basic Search

{
  "query": "artificial intelligence news"
}

Advanced Search

{
  "query": "machine learning",
  "num": 10,
  "lr": "lang_en",
  "gl": "us",
  "dateRestrict": "m1",
  "fileType": "pdf"
}

Site-Specific Search

{
  "query": "typescript tutorial",
  "siteSearch": "stackoverflow.com",
  "siteSearchFilter": "i"
}

Quota Management

The server provides real-time quota information in each response:

{
  "results": [...],
  "metadata": {
    "quotaStatus": {
      "totalUsed": 15,
      "totalLimit": 300,
      "keysStatus": [
        {
          "id": "key_1",
          "used": 15,
          "limit": 100,
          "remaining": 85,
          "active": true
        }
      ]
    }
  }
}

Quota Features

  • Automatic daily reset at midnight UTC
  • Intelligent key rotation when limits are reached
  • Disabled keys automatically reactivate after reset
  • Persistent tracking across server restarts
  • No charges - stops at free tier limits

Error Handling

The server gracefully handles various error scenarios:

  • Quota Exceeded: Automatically rotates to next available key
  • All Keys Exhausted: Returns clear error message with quota status
  • Invalid API Key: Disables the key and continues with others
  • Network Errors: Returns detailed error information

Development

Prerequisites

  • Node.js 18+
  • TypeScript
  • Google Cloud account

Scripts

# Development mode with auto-reload
npm run dev

# Build for production
npm run build

# Start server
npm start

# Setup configuration
npm run setup

Project Structure

google-search-mcp/
├── src/
│   ├── index.ts         # Entry point
│   ├── server.ts        # MCP server setup
│   ├── config.ts        # Configuration wrapper
│   ├── global-config.ts # Global config manager
│   └── tools/
│       └── search.ts    # Search implementation
├── build/               # Compiled JavaScript
├── setup.js             # Interactive setup tool
└── package.json

Contributing

Contributions are welcome! Please:

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Add tests if applicable
  5. Submit a pull request

License

MIT License - see LICENSE file for details

Support

For issues, questions, or contributions, please visit: https://github.com/Fabien-desablens/google-search-mcp

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选