FLUX Image Generator (Black Forest Lab)

An MCP (Model Context Protocol) server for generating images using Black Forest Lab's FLUX model. Uses the latest MCP SDK (v1.7.0).

Features

• Generate images based on text prompts
• Customize image dimensions, prompt upsampling, and safety settings
• Save generated images locally
• Batch image generation from multiple prompts

Prerequisites

• Node.js (v18.0.0 or higher)
• Black Forest Lab API key (get one at https://api.bfl.ml)

Installation

From Source

1. Clone this repository
2. Install dependencies:

npm install

3. Create a .env file based on .env.example and add your Black Forest Lab API key:

BFL_API_KEY=your_api_key_here

4. Build the project:

npm run build

Using npm

npm install -g @modelcontextprotocol/server-flux-image-generator

Usage

Starting the MCP Server

Start the server with:

npm start

For development with auto-recompilation:

npm run watch

Integrating with MCP Clients

To use this server with MCP clients (like Claude), add the following to your client's configuration:

{
  "mcpServers": {
    "flux-image-generator": {
      "command": "mcp-server-flux-image-generator",
      "env": {
        "BFL_API_KEY": "your_api_key_here"
      }
    }
  }
}

Available Tools

generateImage

Generates an image based on a text prompt with customizable settings.

Parameters:

• prompt (string, required): Text description of the image to generate
• width (number, optional, default: 1024): Width of the image in pixels
• height (number, optional, default: 1024): Height of the image in pixels
• promptUpsampling (boolean, optional, default: false): Enhance detail by upsampling the prompt
• seed (number, optional): Random seed for reproducible results
• safetyTolerance (number, optional, default: 3): Content moderation tolerance (1-5)

Example:

{
  "prompt": "A serene lake at sunset with mountains in the background",
  "width": 1024,
  "height": 768,
  "promptUpsampling": true,
  "seed": 12345,
  "safetyTolerance": 3
}

quickImage

A simplified tool for quickly generating images with default settings.

Parameters:

• prompt (string, required): Text description of the image to generate

Example:

{
  "prompt": "A futuristic cityscape with flying cars"
}

batchGenerateImages

Generates multiple images from a list of prompts.

Parameters:

• prompts (array of strings, required): List of text prompts (maximum 10)
• width (number, optional, default: 1024): Width of the images
• height (number, optional, default: 1024): Height of the images

Example:

{
  "prompts": [
    "A serene lake at sunset",
    "A futuristic cityscape",
    "A magical forest with glowing plants"
  ],
  "width": 1024,
  "height": 768
}

Output Format

All tools return responses in this format:

{
  "image_url": "https://storage.example.com/generated_image.jpg",
  "local_path": "/path/to/output/flux_1234567890.png"
}

For errors:

{
  "error": true,
  "message": "Error description"
}

The batch tool returns:

{
  "total": 3,
  "successful": 2,
  "failed": 1,
  "results": [
    {
      "prompt": "A serene lake at sunset",
      "success": true,
      "image_url": "https://storage.example.com/image1.jpg",
      "local_path": "/path/to/output/flux_batch_1234567890_0.png"
    },
    {
      "prompt": "A futuristic cityscape",
      "success": true,
      "image_url": "https://storage.example.com/image2.jpg",
      "local_path": "/path/to/output/flux_batch_1234567890_1.png"
    },
    {
      "prompt": "Prohibited content",
      "success": false,
      "error": "Content policy violation"
    }
  ]
}

License

MIT