📌Overview

Purpose: MCP Flux Studio aims to seamlessly integrate Flux's advanced image generation capabilities into AI coding assistants, enhancing the development workflow.

Overview: MCP Flux Studio serves as a powerful server that bridges AI coding tools with the Flux image generation API. It enables developers to incorporate advanced image creation and manipulation functionalities directly into IDEs like Cursor and Windsurf, thereby streamlining the development process.

Key Features:

• Image Generation: Offers text-to-image capabilities with various models and customizable dimensions, providing users with control over their outputs.

• Image Manipulation: Allows for transformations and enhancements including inpainting and resolution upscaling, making it easy to refine images.

• Advanced Controls: Features methods like edge-based, depth-aware, and pose-guided generation to enhance creative options for developers.

• IDE Integration: Fully compatible with Cursor and Windsurf/Codeium, enabling straightforward invocation of tools via AI assistants and improving productivity.

---

MCP Flux Studio

MCP Flux Studio is a powerful Model Context Protocol (MCP) server that integrates Flux's advanced image generation capabilities into AI coding assistants like Cursor and Windsurf (Codeium).

Overview

MCP Flux Studio enhances the development workflow by bridging AI coding assistants with Flux's image generation API, allowing for seamless integration of image generation capabilities.

Features

• Image Generation

  • Text-to-image generation
  • Support for multiple models
  • Customizable aspect ratios and dimensions

• Image Manipulation

  • Image-to-image transformation
  • Inpainting with masks
  • Resolution upscaling

• Advanced Controls

  • Edge-based and depth-aware generation
  • Pose-guided generation

• IDE Integration

  • Support for Cursor and Windsurf/Codeium

Quick Start

Prerequisites

• Node.js 18+
• Python 3.12+
• Flux API key
• Compatible IDE (Cursor or Windsurf)

Installation

Installing via Smithery

npx -y @smithery/cli install @jmanhype/mcp-flux-studio --client claude

Manual Installation

git clone https://github.com/jmanhype/mcp-flux-studio.git
cd mcp-flux-studio
npm install
npm run build

Basic Configuration

BFL_API_KEY=your_flux_api_key
FLUX_PATH=/path/to/flux/installation

For detailed setup instructions, see our 1.

Documentation

• 1
• 1
• 1
• 1

IDE Integration

Cursor (v0.45.7+)

1. Configuration: Configure via Settings > Features > MCP
2. Usage: Tools available to Cursor's AI assistant require user approval for invocation.

Windsurf/Codeium (Wave 3+)

1. Configuration: Edit ~/.codeium/windsurf/mcp_config.json
2. Usage: Access tools through Cascade's MCP toolbar.

For detailed IDE-specific setup instructions, see the 1.

Usage

Generate

Generate an image from a text prompt.

{
  "prompt": "A photorealistic cat",
  "model": "flux.1.1-pro",
  "aspect_ratio": "1:1",
  "output": "generated.jpg"
}

Img2Img

Generate an image using another image as reference.

{
  "image": "input.jpg",
  "prompt": "Convert to oil painting",
  "model": "flux.1.1-pro",
  "strength": 0.85,
  "output": "output.jpg"
}

Inpaint

Inpaint using a mask.

{
  "image": "input.jpg",
  "prompt": "Add flowers",
  "mask_shape": "circle",
  "position": "center",
  "output": "inpainted.jpg"
}

Control

Generate an image using structural control.

{
  "type": "canny",
  "image": "control.jpg",
  "prompt": "A realistic photo",
  "output": "controlled.jpg"
}

Development

Project Structure

flux-mcp-server/
├── src/
│   ├── index.ts          # Main server implementation
│   └── types.ts          # TypeScript type definitions
├── tests/
│   └── server.test.ts    # Server tests
├── docs/
│   ├── API.md           # API documentation
│   └── CONTRIBUTING.md  # Contribution guidelines
├── examples/
│   ├── generate.json    # Example tool usage
│   └── config.json      # Example configuration
├── package.json
├── tsconfig.json
└── README.md

Running Tests

npm test

Building

npm run build

Contributing

Please read 1 for details on contributing.

License

This project is licensed under the MIT License.