Gemini Image MCP Server

🚀 Gemini Image MCP Server

An image generation and editing service based on Google Gemini, providing AI image processing capabilities through the MCP protocol.

🚀 Quick Start

✨ Features

• Text-to-image generation (generate_image)
• Image editing and modification (edit_image)
• Support for PNG format
• Automated file path management

📦 Installation

# Install dependencies
uv pip install -r requirements.txt

# Set environment variables (Linux/macOS)
export GEMINI_API_KEY=your_api_key_here

# Windows settings:
setx GEMINI_API_KEY "your_api_key_here"

💻 Usage Examples

Basic Usage

# generate_image API example
@mcp.tool()
async def generate_image(prompt: str, save_path: str = None) -> str:
    """
    :param prompt: Image generation prompt (better in English)
    :param save_path: Optional save path (default: generated-images/ directory)
    :return: Absolute path of the generated image
    """

# edit_image API example
@mcp.tool()
async def edit_image(prompt: str, image_path: str, save_path: str = None) -> str:
    """
    :param prompt: Image editing instruction (supports Chinese)
    :param image_path: Path of the original image
    :param save_path: Optional save path, default is generated in the generated-images/ directory
    :return: Absolute path of the edited image
    """

📚 Documentation

MCP Configuration

Add the following to mcp.json:

{
    "mcpServers": {
        "gemini-image-mcp-server": {
            "command": "uv",
            "args": [
                "--directory",
                "/ABSOLUTE/PATH/TO/PARENT/FOLDER/",
                "run",
                "server.py"
            ]
        }
    }
}

⚠️ Important Note

1. Ensure that the paths in the configuration are replaced with absolute paths.
2. The API key must be set via environment variables.
3. It is recommended to run the example code for initial use.
4. Generated images are saved in the generated-images directory by default.