Freepik MCP

How to Install and Use Freepik MCP

If you want to connect your AI assistant directly to Freepik's APIs to generate, search, and manage visual content, the Freepik MCP (Model Context Protocol) is the right tool. It makes it easy to work with Freepik inside your AI workflow. Here's a simple guide to get it installed and running.

Step 1: Prepare Your System

Before we start, make sure your system is ready. You need to have Python 3.12 or higher installed on your computer. Also, Freepik MCP uses a special tool called uv to manage dependencies. If you don’t have it yet, you can install it following the instructions here: uv installation guide.

Lastly, you need a Freepik API key to access their services. You can get your free API key by signing up at freepik.com/api.

Step 2: Clone and Navigate to the Freepik MCP Project

Now that your system is ready, let's get the Freepik MCP files on your computer. Open your command line or terminal and run:

git clone https://github.com/freepik-company/freepik-mcp.git
cd freepik-mcp

This command downloads the Freepik MCP project and moves you into its folder so you can work with it.

Step 3: Install Required Dependencies

Freepik MCP needs some special Python libraries to work properly. Luckily, it comes with a Makefile that helps you install them easily. Run:

make install

This command will download and set up everything Freepik MCP needs to run.

Step 4: Configure Your Freepik API Key

To connect Freepik MCP with Freepik's services, you must provide your API key securely. Create a .env file with the key by running:

echo "FREEPIK_API_KEY=your_api_key_here" > .env

Make sure to replace your_api_key_here with the actual key you got from the Freepik website.

Step 5: Run Freepik MCP Server

Now you are ready to start the Freepik MCP server. This server allows your AI assistants like Claude or Cursor to communicate with Freepik APIs. For development and auto-reload, use:

make dev

If you want to run it in a stable production mode, use:

make run

Step 6: Setup for AI Assistants

If you want to connect Freepik MCP with AI assistants such as Claude Desktop or Cursor (on Linux), add this configuration to their respective config.json file. Remember to replace the path and API key with your actual values:

{
  "mcpServers": {
    "freepik-fastmcp": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/FULL/PATH/TO/freepik-mcp",
        "main.py"
      ],
      "env": {
        "FREEPIK_API_KEY": "your_actual_api_key_here"
      }
    }
  }
}

To find your full path, you can run pwd in the terminal inside the freepik-mcp folder and copy the printed path.

Note for Windows Users: You will need to use WSL (Windows Subsystem for Linux) to run this MCP server.

Conclusion

By following these simple steps, you can install and use the Freepik MCP tool to integrate Freepik’s powerful image and icon services directly into your AI assistant workflows. This setup allows you to search, generate, and manage visual content quickly and effectively, making your AI projects much more creative and visual.

Remember to check the Freepik API documentation if you want to explore all the options Freepik MCP can offer!