Image Generation (Flux Schnell)

🚀 Image Generation MCP Server

An implementation of an MCP (Model Context Protocol) server for generating images using the black-forest-labs/flux-schnell model from Replicate. It's ideally used with Cursor's MCP feature but can also work with any other MCP clients.

✨ Features

• Generate images from text prompts.
• Configurable image parameters (resolution, aspect ratio, quality).
• Save generated images to a specified directory.
• Fully compliant with the MCP protocol.
• Error handling and validation.

📦 Prerequisites

• Node.js 16+
• Replicate API token
• MCP SDK for TypeScript

🛠️ Installation

1. Clone the repository.

2. Install dependencies:

   npm install
   

3. Update the apiToken constant directly in src/imageService.ts with your Replicate API token:

   // Add your API token directly here as environment variables are not easily set in Cursor
   const apiToken = "your-replicate-api-token-here";
   

   ⚠️ Important Note

   If using Claude, create a .env file in the root directory and set the API token as follows:

   REPLICATE_API_TOKEN=your-replicate-api-token-here
   

   Then build the project:

   npm run build
   

💻 Usage Examples

Basic Usage

When used with Cursor:

1. Go to Settings.
2. Select Features.
3. Scroll to the bottom to "MCP Server".
4. Click "Add New MCP Server".
5. Set the type to "Command".
6. Set the command to: node ./path/to/dist/server.js

API Parameters

Example Request

{
  "prompt": "Black forest cake with 'FLUX SCHNELL' written on it",
  "output_dir": "/var/output/images",
  "filename": "black_forest_cake",
  "output_format": "webp",
  "go_fast": true,
  "megapixels": "1",
  "num_outputs": 2,
  "aspect_ratio": "1:1"
}

Example Response

{
  "status": "success",
  "message": "Images have been generated and saved to the specified directory.",
  "data": {
    "filename": "black_forest_cake.png",
    "size": "2.5 MB",
    "dimensions": "1920x1080"
  }
}

🔧 Technical Details

Error Handling

• Input Errors: If the prompt or parameters are invalid, a validation error will be returned.
• API Errors: If the call to the Replicate API fails, relevant error information will be returned.
• Saving Errors: If the images cannot be saved to the specified directory, a storage error will be returned.

📄 License

This code is licensed under the MIT License.