MCP Server Archived reference

EverArt

MCP server that generates images through EverArt's API from Claude Desktop or VS Code, with several model options and a single generate_image tool.

  • 304 GitHub stars
docker build -t mcp/everart -f src/everart/Dockerfile .
EverArt preview image

What it is

EverArt MCP Server is an image generation server for Claude Desktop that uses EverArt's API. It exposes a single tool, generate_image, which supports multiple models (FLUX1.1, FLUX1.1-ultra, SD3.5, Recraft-Real, Recraft-Vector). The tool opens the result in the default browser and returns the image URL. It can be run via NPX or Docker and configured for Claude Desktop or VS Code.

Who it's for

  • Claude Desktop users who want to generate images from chat
  • VS Code users who want image generation through MCP
  • Developers evaluating EverArt models through an MCP client

Requirements

Requirements

  • An EverArt API key, set as the EVERART_API_KEY environment variable
  • Claude Desktop or VS Code as the MCP client
  • Node.js/npm (for the NPX or npm install route) or Docker

Setup

  1. Install and set API key

    Install dependencies and export your EverArt API key.

    bash
    npm install
    export EVERART_API_KEY=your_key_here
  2. Claude Desktop config (NPX)

    Add this to the Claude Desktop config.

    json
    {
      "mcpServers": {
        "everart": {
          "command": "npx",
          "args": ["-y", "@modelcontextprotocol/server-everart"],
          "env": {
            "EVERART_API_KEY": "your_key_here"
          }
        }
      }
    }
  3. Claude Desktop config (Docker)

    Alternatively, run the server through Docker using the mcp/everart image.

    json
    {
      "mcpServers": {
        "everart": {
          "command": "docker",
          "args": ["run", "-i", "--rm", "-e", "EVERART_API_KEY", "mcp/everart"],
          "env": {
            "EVERART_API_KEY": "your_key_here"
          }
        }
      }
    }
  4. Build the Docker image

    Build the image locally if you want to use the Docker option.

    sh
    docker build -t mcp/everart -f src/everart/Dockerfile .

Examples

Call generate_image

javascript
javascript
const result = await client.callTool({
  name: "generate_image",
  arguments: {
    prompt: "A cat sitting elegantly",
    model: "7000",
    image_count: 1
  }
});

What it does: Generates one image of a cat using model 7000 (Recraft-Real).

generate_image parameters

typescript
typescript
{
  prompt: string,       // Image description
  model?: string,       // Model ID (default: "207910310772879360")
  image_count?: number  // Number of images (default: 1)
}

What it does: Only prompt is required; model and image_count are optional.

Response format

text
text
Image generated successfully!
The image has been opened in your default browser.

Generation details:
- Model: 7000
- Prompt: "A cat sitting elegantly"
- Image URL: https://storage.googleapis.com/... 

You can also click the URL above to view the image again.

What it does: The tool opens the image in the browser and returns its URL with generation details.

Pros & cons

Pros

  • Pro:Offers several models, including FLUX1.1, FLUX1.1-ultra, SD3.5, Recraft-Real and Recraft-Vector
  • Pro:Simple single-tool interface with only one required parameter (prompt)
  • Pro:Supports both NPX and Docker installation, with configs for Claude Desktop and VS Code
  • Pro:Returns the image URL and opens the result in the browser

Cons

  • Con:All images are generated at a fixed 1024x1024 size
  • Con:Requires an EverArt API key
  • Con:Repository is in the archived servers repo, as the URL indicates

Images