Gemini Search

🚀 Libra Search MCP Server

An MCP server that uses the Gemini API and Google Search to generate responses based on the latest information.

⚠️ Important Note

This MCP server cannot run independently. It needs to be used in conjunction with an AI assistant like Cline. After loading this project into an AI assistant, the Gemini search function can be enabled.

🚀 Quick Start

The Libra Search MCP Server is designed to generate responses using the Gemini API and Google Search. To get started, follow the installation and setup steps below.

✨ Features

Tools

• search - Generate responses using Gemini 2.0 and Google Search
  • It takes a query as input and returns the response from Gemini along with relevant search results.

📦 Installation

1. Install dependencies:

npm install

2. Build the project:

npm run build

3. Set up environment variables: Create a .env file in the project root directory with the following content:

GEMINI_API_KEY=your_api_key_here

Note: You can obtain the Gemini API key at Google AI Studio.

💻 Usage Examples

Development

For automatic builds during development:

npm run watch

Installation Guide

When using it with Claude Desktop, add the following configuration:

Windows system path: %APPDATA%/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "gemini": {
      "command": "node",
      "args": ["path/to/gemini-search-server/build/index.js"],
      "env": {
        "GEMINI_API_KEY": "your_api_key_here"
      }
    }
  }
}

Debugging

Since the MCP server communicates via stdio, we recommend using MCP Inspector for debugging:

npm run inspector

The Inspector provides a URL to access the debugging tool in your browser.

📄 License

The code of this project is released under the MIT License. However, please note that this project uses the Google Gemini API and must comply with its Terms of Service. When using this MCP server, you must adhere to both the MIT License and Google's Terms of Service.