MCP Google Gemini Image Generation

MCP server for generating images using Google's Gemini API with Context7 integration.

Features

• Generate images from text prompts using Google Gemini API
• Save generated images to a specified directory
• Configure different Gemini models for image generation
• Context7 integration with sequential thinking capability
• Supports both programmatic use and CLI interface

Installation

npm install -g mcp-google-gemini-image-generation

Or use it directly with npx:

npx mcp-google-gemini-image-generation

Configuration

Create a .env file in your project directory (you can copy from .env.example):

# Google Gemini API Key
GEMINI_API_KEY=your_api_key_here

# Output directory for images
OUTPUT_DIR=./output

# Default model for image generation
DEFAULT_MODEL=gemini-2.0-flash-preview-image-generation

Usage

Using in VS Code with MCP Extension

1. Make sure you have the MCP Extension installed in VS Code
2. Add the server to your MCP configuration:

{
  "servers": {
    "gemini-image-generation": {
      "command": "npx",
      "args": ["-y", "mcp-google-gemini-image-generation"]
    }
  }
}

3. Use it in VS Code by providing context:

prompt: "Create an image of a dragon flying over a magical forest"
outputPath: "./images"
outputFilename: "dragon.png"
model: "gemini-2.0-flash-preview-image-generation"

Using with CLI

npx mcp-google-gemini-image-generation --api-key YOUR_API_KEY --output-dir ./images --model gemini-2.0-flash-preview-image-generation

Using Sequential Thinking

This MCP server includes a sequential thinking tool that helps break down complex image generation tasks:

{
  "thoughtNumber": 1,
  "totalThoughts": 3,
  "thought": "First, let's consider the key elements needed in this fantasy scene...",
  "nextThoughtNeeded": true
}

API Reference

Context Parameters

Parameter	Description	Required	Default
prompt	The text prompt to generate an image from	Yes	-
model	The Gemini model to use	No	gemini-2.0-flash-preview-image-generation
outputPath	Directory to save the generated image	No	./output
outputFilename	Filename for the generated image	No	gemini-image-[uuid].png

Tools

generate-image

Generates an image based on text prompt.

Parameters:

• prompt (required): Text description of the image to generate
• model: Name of the Gemini model to use
• outputPath: Directory to save the image
• outputFilename: Custom filename for the generated image

sequential-thinking

Enables step-by-step complex reasoning for image generation tasks.

Parameters:

• thought (required): Current thinking step
• nextThoughtNeeded (required): Whether another thought step is needed
• thoughtNumber (required): Current thought number
• totalThoughts (required): Estimated total thoughts needed
• isRevision: Whether this revises previous thinking
• revisesThought: Which thought is being reconsidered
• branchFromThought: Branching point thought number
• branchId: Branch identifier
• needsMoreThoughts: If more thoughts are needed

License

ISC