gemini-mcp-server

A TypeScript implementation of a Model Context Protocol (MCP) server that integrates with Google's Gemini model using direct API calls.

Features

• Uses direct calls to the Gemini API (no deprecated SDK)
• Supports the latest Gemini 2.0 Flash model
• Implements MCP protocol for seamless integration with Claude
• Maintains conversation context for natural interactions

MCP Tools

generate_text

From server: gemini

Generate text using Gemini model with configurable parameters.

Prerequisites

• Node.js 18 or higher
• Google Gemini API key
• TypeScript
• Claude Desktop app

Installation

1. Clone the repository:

git clone https://github.com/YOUR-USERNAME/gemini-mcp-server.git
cd gemini-mcp-server

2. Install dependencies:

npm install

3. Build:

npm run build

Claude Desktop Integration

To use this server with Claude Desktop:

1. Open Claude Desktop
2. Go to Settings > Developer
3. Click "Edit Config"
4. Add the following configuration:

"gemini": {
  "command": "node",
  "args": ["path\\to\\dist\\gemini_mcp_server.js"],
  "env": {
    "GEMINI_API_KEY": "gemini_api_key"
  },
  "cwd": "path\\to\\gemini-mcp-server"
}

Replace:

• /path/to/gemini-mcp-server with the absolute path to your repository
• your_api_key_here with your actual Google Gemini API key

The server will now be available in Claude Desktop's MCP server list.

API Implementation

This server uses direct HTTP requests to the Gemini API endpoint. The API request format follows Google's official documentation:

curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key=GEMINI_API_KEY" \
-H 'Content-Type: application/json' \
-X POST \
-d '{
  "contents": [{
    "parts":[{"text": "Explain how AI works"}]
    }]
   }'

Testing

You can test the direct API implementation using the example script:

node dist/example_direct_api.js "Your prompt here"

License

MIT