Generate Image MCP Server

🚀 Image Generation MCP Server

This project implements a Model Context Protocol (MCP) server based on the OpenAI DALL-E model API calls.

This project is adapted from https://github.com/nicekate/flux-dev-mcp. Thanks to the original author for their contribution. This project uses a forwarding API service that is fully compatible with the OpenAI interface, supporting multiple large language model APIs such as OpenAI and Anthropic, as well as text-to-image generation model APIs like MidJourney and Flux. The latest model, claude-3-7-sonnet, is now online!

Register using my referral link https://xiaohumini.site/register?aff=Zbtb and get $0.4 in free credits!

✨ Features

• Based on the OpenAI DALL-E model API calls.
• Adapted from an existing project on GitHub.
• Uses a forwarding API service fully compatible with the OpenAI interface.
• Supports multiple large language model APIs and text-to-image generation model APIs.
• The latest model claude-3-7-sonnet is available.

📦 Installation

Prerequisites

• Node.js 16+
• npm or yarn
• Windows or MacOS operating system

Install Dependencies

# Universal command for Windows/MacOS
npm install

Build the Project

# Universal command for Windows/MacOS
npm run build

Development Mode

Start the development mode with automatic rebuild:

# Universal command for Windows/MacOS
npm run watch

💻 Usage Examples

Cline/ROO CODE Plugin Configuration

In the Cline/ROO CODE plugin, you need to configure the startup command for the MCP server. Please follow these steps:

Windows Configuration Example

{
  "mcpServers": {
    "genimg-mcp": {
      "command": "node",
      "arg": ["C:\\path\\to\\genimg-mcp\\dist\\index.js"],
      "env": {
        "OPENAI_API_KEY": "sk-xxxxx",
        "OPENAI_MODEL_ID": "dall-e-3"  # Optional
      }
    }
  }
}

MacOS Configuration Example

{
  "mcpServers": {
    "genimg-mcp": {
      "command": "/path/to/genimg-mcp/dist/index.js",
      "env": {
        "OPENAI_API_KEY": "sk-xxxxx",
        "OPENAI_MODEL_ID": "dall-e-3"  # Optional
      }
    }
  }
}

⚠️ Important Note

• Use double backslashes \\ or single forward slashes / for Windows paths.
• Use single forward slashes / for MacOS paths.
• Make sure the paths point to the actual installation locations.

📚 Documentation

Debugging Guide

Since the MCP server communicates via standard input/output (stdio), debugging can be challenging. We recommend using the MCP Inspector tool:

# Universal command for Windows/MacOS
npm run inspector

The Inspector will provide a browser access address for the debugging tool.

📧 Contact

For support, please contact 1813345945@qq.com.

📄 License

This project is licensed under the MIT license. See the LICENSE file for details.

🔧 Technical Details

Technology Stack

• TypeScript
• Node.js
• Duannaoyun API