Gemini Image Generation MCP Server
This is a Model Context Protocol (MCP) server that provides image generation capabilities using Google's Gemini 2 API.
Quick Start
-
Get Gemini API Key
- Visit Google AI Studio
- Create a new API key
-
Configure Claude Desktop
- Locate your config file:
Mac: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json Linux: ~/.config/Claude/claude_desktop_config.json - Add Gemini configuration:
{ "mcpServers": { "gemini-imagen": { "command": "npx", "args": ["-y", "github:sanxfxteam/gemini-mcp-server"], "env": { "GEMINI_API_KEY": "your_api_key_here" } } } }
- Locate your config file:
-
Restart Claude Desktop
Setup
- Install dependencies:
npm install
- Set up your environment variables:
Create a
.envfile in the root directory and add your Google API key:
GEMINI_API_KEY=your_api_key_here
Usage
Run the server:
npm start
To test
npx @modelcontextprotocol/inspector npm run start
Available Tools
generateImage
Generates images using Gemini 2's experimental image generation API.
Parameters:
prompt(string, required): The description of the image you want to generatenumSamples(number, optional, default: 4): Number of images to generateaspectRatio(string, optional, default: '1:1'): Aspect ratio of the generated imagespersonGeneration(string, optional, default: 'ALLOW_ADULT'): Person generation settings
Example MCP request:
{
"tool": "generateImage",
"params": {
"prompt": "A serene mountain landscape at sunset",
"numSamples": 2,
"aspectRatio": "16:9"
}
}
Notes
- This server uses the experimental image generation feature of Gemini 2
- Make sure you have appropriate access and API keys from Google
- The server communicates using the Model Context Protocol over stdio