Image Gen MCP

🚀 Image Generation MCP Server

An MCP server that provides text-to-image generation functionality, utilizing the Stable Diffusion WebUI API (ForgeUI/AUTOMATIC-1111).

🚀 Quick Start

📦 Installation

Prerequisites

• Node.js
• Access to a Stable Diffusion WebUI instance with API enabled
• The --api flag must be used when starting the WebUI

Setup

1. Clone the repository:

git clone https://github.com/Ichigo3766/image-gen-mcp.git
cd image-gen-mcp

2. Install dependencies:

npm install

3. Build the server:

npm run build

4. Add server configuration to environment variables:

{
  "mcpServers": {
    "image-gen": {
      "command": "node",
      "args": [],
      "env": {
        "SD_WEBUI_URL": "http://localhost:7860",
        "SD_API_KEY": "your_api_key", // Optional, if the WebUI has API key enabled
        "SD_MODEL_NAME": "Model Name" // Optional, defaults to the default model in the WebUI
      }
    }
  }
}

✨ Features

Text-to-Image Generation

Use the /api/generate-image endpoint to generate an image. The request body should contain the following:

{
  "prompt": "Input prompt",
  "negative_prompt": "Excluded prompt", // Optional, defaults to empty
  "width": 512,
  "height": 512,
  "samples": 1,
  "num_inference_steps": 20,
  "safety_checker": true, // Toggle safety check to prevent inappropriate content generation, enabled by default
  "enhance_prompt": true, // Enable prompt enhancement, enabled by default
  "seed": null // Random seed, optional, defaults to a random value
}

Image Upscaling

Use the /api/upscale-images endpoint to upscale images. The request body should contain the following:

{
  "images": ["image1.jpg", "image2.png"], // Array of file paths of images to be upscaled
  "resize_mode": 0, // 0 for proportional scaling, 1 for fixed size, defaults to value from environment variables
  "upscaling_resize": 4, // Upscaling ratio, used when resize_mode=0, defaults to value from environment variables
  "upscaling_resize_w": 1024, // Target width (pixels), used when resize_mode=1, defaults to value from environment variables
  "upscaling_resize_h": 1024, // Target height (pixels), used when resize_mode=1, defaults to value from environment variables
  "upscaler_1": "RealESRGAN_x4Plus", // Primary upscaler model, defaults to value from environment variables
  "upscaler_2": null, // Secondary upscaler model, defaults to null
  "output_path": "upscaled_images" // Output directory for upscaled images, defaults to current directory/upscaled-images
}

💻 Development

With automatic rebuild functionality:

npm run watch

🛠️ Error Handling

Common issues and solutions:

1. Ensure that your Stable Diffusion WebUI is running with the --api flag.
2. Check if the WebUI URL is accessible from the MCP server.
3. If authentication is used, ensure that the credentials are correct.
4. Verify that the output directory exists and has write permissions.
5. When upscaling images, ensure that the input image files exist and are readable.

📄 License

This MCP server is licensed under the MIT License. This means you are free to use, modify, and distribute the software, but you must comply with the terms and conditions of the MIT License. For more details, please refer to the LICENSE file in the project repository.