DALL-E

This MCP server lets you generate and edit images using OpenAI’s DALL-E API through a simple, scriptable interface. You can create new images from prompts, edit existing images, or generate variations, all orchestrated through MCP clients for seamless conversations and workflows.

How to use

You interact with the server through an MCP client to generate, edit, or vary images with DALL-E. Start by starting the local server, then issue tool requests from your MCP client. The server supports three main actions: generate images from prompts, edit existing images with a prompt, and create variations of an image. You can also validate your OpenAI API key to ensure the server can access the OpenAI services.

How to install

Prerequisites: you need Node.js and npm installed on your system. Ensure you have an OpenAI API key ready.

Step 1: Clone the project and navigate into the directory.

# Clone the repository
git clone https://github.com/Garoth/dalle-mcp.git
cd dalle-mcp

Step 2: Install dependencies.

npm install

Step 3: Build the project.

npm run build

Step 4: Run the server locally.

node build/index.js

Step 5: Configure your MCP client (example for Cline is shown below). You will provide your OpenAI API key and an optional default save directory.

The client configuration for Cline can point to the local server using a stdio setup. The following example shows how to wire the built server so your MCP client can launch it and pass environment variables.

{
  "mcpServers": {
    "dalle-mcp": {
      "command": "node",
      "args": ["/path/to/dalle-mcp-server/build/index.js"],
      "env": {
        "OPENAI_API_KEY": "your-api-key-here",
        "SAVE_DIR": "/path/to/save/directory"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

Additional notes for usage

When using this DALL-E MCP server with a client like Cline, consider saving generated images in your current workspace. Set the saveDir parameter to match your working directory to help the client locate and display newly generated images in conversations.

Configuration for MCP clients

You can wire the server to run under MCP clients by providing the built index file path and your API key in the environment. This ensures the client can launch the server and send prompts to generate images, apply edits, or create variations.

Security and keys

Protect your OpenAI API key. Do not share the key in public configurations. Rotate keys if you suspect they have been compromised.

Troubleshooting and notes

If image edits sometimes do not apply as expected with the edit function, try using the variation tool as an alternative. The edit API may occasionally return the original image due to known issues.

Available tools

generate_image

Generate an image from a text prompt using DALL-E with options for model, size, quality, and style.

edit_image

Edit an existing image based on a text prompt. Currently the DALL-E 2 edit path has known reliability considerations.

create_variation

Create variations of an existing image using DALL-E, enabling multiple outputs from a single input image.

validate_key

Validate that the provided OpenAI API key is usable for accessing DALL-E features.