OpenAI Image Generator MCP server

This MCP server provides a powerful interface to OpenAI's GPT-4o and GPT-Image-1 APIs, allowing you to generate and edit images through any MCP-compatible client such as Claude Desktop, Cursor, VSCode, or Windsurf.

Installation

To install the OpenAI GPT Image MCP server:

git clone https://github.com/SureScaleAI/openai-gpt-image-mcp.git
cd openai-gpt-image-mcp
yarn install
yarn build

Configuration

Add the server to your Claude Desktop or VSCode (including Cursor/Windsurf) configuration:

{
  "mcpServers": {
    "openai-gpt-image-mcp": {
      "command": "node",
      "args": ["/absolute/path/to/dist/index.js"],
      "env": { "OPENAI_API_KEY": "sk-..." }
    }
  }
}

Be sure to replace /absolute/path/to/dist/index.js with the actual path and add your OpenAI API key.

Features

Generate Images

Use the create-image function to generate images from text prompts with various options:

• Size customization
• Quality settings
• Background options
• Generate up to 10 images at once

Edit Images

The edit-image function allows you to modify existing images:

• Inpainting (modify specific areas)
• Outpainting (extend beyond the original image)
• Compositing (combine elements)
• Support for both file paths and base64 input

Usage Examples

The server provides two main functions that can be called from any MCP-compatible client:

Creating Images

create-image
prompt: A serene mountain landscape at sunset
size: 1024x1024
quality: hd
n: 1
file_output: /path/to/save/image.png

Editing Images

edit-image
image: /path/to/original_image.png
prompt: Add a small cabin in the foreground
mask: /path/to/mask_image.png
file_output: /path/to/save/edited_image.png

Troubleshooting

• Ensure your OPENAI_API_KEY is valid and has image API access
• Verify you have a verified OpenAI organization
• After verification, allow 15-20 minutes for image API access to activate
• File paths must be absolute
• Ensure output directories are writable
• Check image file extensions and formats if you encounter file type errors

Handling Large Files

• MCP clients have a 1MB payload limit for tool responses
• For images exceeding 1MB, the tool automatically saves to disk and returns file paths
• Set the MCP_HF_WORK_DIR environment variable to control where large images are saved
• For large or production images, always use the file output option

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.