Recraft

MCP (modelcontextprotocol) server implementation for Recraft API
  • typescript

46

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "recraft-ai-mcp-recraft-server": {
      "command": "npx",
      "args": [
        "-y",
        "@recraft-ai/mcp-recraft-server@latest"
      ],
      "env": {
        "RECRAFT_API_KEY": "<YOUR_RECRAFT_API_KEY>",
        "IMAGE_STORAGE_DIRECTORY": "<YOUR_IMAGE_STORAGE_DIRECTORY>",
        "RECRAFT_REMOTE_RESULTS_STORAGE": "<YOUR_REMOTE_RESULTS_STORAGE_INDICATOR>"
      }
    }
  }
}

You connect this MCP server to your MCP client to generate high-quality raster and vector images using Recraft. You can perform image generation, editing, style creation, vectorization, background operations, and upscaling all through MCP-enabled workflows, combining multiple tools in a single pipeline.

How to use

You interact with the server through an MCP client to request image generation and manipulation. Start by ensuring your client is configured to access the Recraft MCP server, then issue tool calls such as generating images from prompts, vectorizing rasters, or applying backgrounds and upscales. Each tool returns results you can download or reference via URLs when remote storage is enabled.

How to install

Prerequisites: ensure Node.js is installed on your machine. If you don’t have Node, download it from the official site and install it.

Install via NPM or build from source using the commands below. Follow the steps that match your setup.

{
  "mcpServers": {
    "recraft": {
      "command": "npx",
      "args": [
        "-y",
        "@recraft-ai/mcp-recraft-server@latest"
      ],
      "env": {
        "RECRAFT_API_KEY": "<YOUR_RECRAFT_API_KEY>",
        "IMAGE_STORAGE_DIRECTORY": "<YOUR_IMAGE_STORAGE_DIRECTORY>",
        "RECRAFT_REMOTE_RESULTS_STORAGE": "<YOUR_REMOTE_RESULTS_STORAGE_INDICATOR>"
      }
    }
  }
}
# From NPM (if you prefer using your existing setup)
# This snippet shows the config you would place in your Claude Desktop setup

From source

Clone the Recraft MCP server repository, install dependencies, and build the project.

git clone https://github.com/recraft-ai/mcp-recraft-server.git
cd mcp-recraft-server
npm install
npm run build

Configure your MCP client to start the server from the built dist file.

{
  "mcpServers": {
    "recraft_local": {
      "command": "node",
      "args": ["<ABSOLUTE_PATH_TO_CLONED_DIRECTORY>/dist/index.js"],
      "env": {
        "RECRAFT_API_KEY": "<YOUR_RECRAFT_API_KEY>",
        "IMAGE_STORAGE_DIRECTORY": "<YOUR_IMAGE_STORAGE_DIRECTORY>",
        "RECRAFT_REMOTE_RESULTS_STORAGE": "<YOUR_REMOTE_RESULTS_STORAGE_INDICATOR>"
      }
    }
  }
}

Notes on prerequisites and storage

You must provide a Recraft API key from your Recraft account. This key is used to authorize image generation and processing requests.

You can choose to store generated images locally in a directory you specify with IMAGE_STORAGE_DIRECTORY or opt to store all results remotely by setting RECRAFT_REMOTE_RESULTS_STORAGE to 1. If remote storage is enabled, the local storage path is ignored.

Security and usage tips

Keep your API key confidential. Do not share your key in public configurations. Use environment variables to manage secrets securely.

When using remote storage, ensure your access permissions and storage policies align with your project requirements. Regularly rotate keys and monitor usage.

Troubleshooting

If you encounter issues during installation, verify you have the latest version of your MCP client and that Node is installed correctly. Recheck the environment variables you provided and confirm your API key is valid.

Example tool usage overview

You can call tools like generate_image, vectorize_image, and remove_background through your MCP client. Each tool accepts specific parameters such as prompts, styles, image inputs, and size. Results can be stored remotely or locally depending on your configuration.

Available tools

generate_image

Generates raster/vector images from a text prompt with optional style, size, model, and quantity.

create_style

Creates a style from a list of images and returns a reusable style for future generations.

vectorize_image

Converts a raster image into a vector representation.

image_to_image

Generates new images from an existing image and a prompt, with optional similarity strength, style, and model.

remove_background

Removes the background from an image.

replace_background

Generates a new background for an image from a prompt with style and size options.

crisp_upscale

Performs a high-quality upscale of an image.

creative_upscale

Performs an upscale with creative enhancements.

get_user

Retrieves information about the user and remaining balance.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational