Unity MCP server

Server + Plugin for Unity Editor and Unity game. The Plugin allows to connect to MCP clients like Claude Desktop or others.
Back to servers
Setup instructions
Provider
Ivan Murzak
Release date
Apr 15, 2025
Language
C#
Stats
570 stars

The Unity MCP server acts as a bridge between Large Language Models (LLMs) and Unity, allowing for AI-powered assistance in game development. This server enables both editor and runtime communication with AI models.

Setting Up MCP Server

The MCP server provides a connection between your Unity project and AI assistants like Claude, Cursor, or Windsurf through the Model Context Protocol.

Installation Options

Docker Container

# Basic HTTP transport setup
docker run -p 8080:8080 ivanmurzakdev/unity-mcp-server

# STDIO transport setup
docker run -t -e UNITY_MCP_CLIENT_TRANSPORT=stdio -p 8080:8080 ivanmurzakdev/unity-mcp-server

# Custom port configuration
docker run -e UNITY_MCP_PORT=123 -p 123:123 ivanmurzakdev/unity-mcp-server

Binary Executable

You can also run the MCP server directly using pre-compiled binaries for your specific platform:

./unity-mcp-server --port 8080 --plugin-timeout 10000 --client-transport stdio

Configuration Variables

The server supports configuration through both environment variables and command line arguments:

Environment Variable Command Line Arg Description
UNITY_MCP_PORT --port Connection port (default: 8080)
UNITY_MCP_PLUGIN_TIMEOUT --plugin-timeout Connection timeout in ms (default: 10000)
UNITY_MCP_CLIENT_TRANSPORT --client-transport Transport type: stdio or http (default: http)

Connecting MCP Clients

After setting up the server, you need to configure your MCP client to connect to it. Different clients have different configuration methods:

HTTP Transport Configuration

For HTTP transport, your MCP client configuration should include:

{
  "mcpServers": {
    "Unity-MCP": {
      "url": "http://localhost:8080"
    }
  }
}

STDIO Transport Configuration

For STDIO transport, your MCP client needs to launch the server:

{
  "mcpServers": {
    "Unity-MCP": {
      "command": "docker",
      "args": [
        "run",
        "-t",
        "-e",
        "UNITY_MCP_CLIENT_TRANSPORT=stdio",
        "-p",
        "8080:8080",
        "ivanmurzakdev/unity-mcp-server"
      ]
    }
  }
}

Advanced Usage

Custom MCP Tools

You can extend the MCP server's capabilities by creating custom tools for your specific needs:

[McpPluginToolType]
public class Tool_GameObject
{
    [McpPluginTool
    (
        "MyCustomTask",
        Title = "Create a new GameObject"
    )]
    [Description("Explain here to LLM what this tool does.")]
    public string CustomTask
    (
        [Description("Explain to LLM what this parameter is.")]
        string inputData
    )
    {
        // Background thread operations

        return MainThread.Instance.Run(() =>
        {
            // Main thread operations (for Unity API access)

            return $"[Success] Operation completed.";
        });
    }
}

Runtime Integration

To use MCP within your running game:

// Initialize the system
UnityMcpPlugin.BuildAndStart();

// Start active connection with retry to Unity-MCP-Server
UnityMcpPlugin.Connect();

// When finished, disconnect
UnityMcpPlugin.Disconnect();

Common Use Cases

  • Game AI Development: Create AI-driven NPCs or game systems
  • Automated Asset Creation: Generate and configure game assets
  • Code Generation: Get AI to write or modify code for your project
  • Dynamic Debugging: Use AI to help identify and fix issues at runtime
  • Procedural Content: Create AI-powered procedural content generation

The MCP server provides a flexible foundation for integrating AI capabilities into both your development workflow and your games themselves.

How to install this MCP server

For Claude Code

To add this MCP server to Claude Code, run this command in your terminal:

claude mcp add-json "unity-mcp" '{"command":"dotnet","args":["run","--project","path/to/unity-mcp"]}'

See the official Claude Code MCP documentation for more details.

For 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 > Tools & Integrations and click "New MCP Server".

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

{
    "mcpServers": {
        "unity-mcp": {
            "command": "dotnet",
            "args": [
                "run",
                "--project",
                "path/to/unity-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 explicitly ask the agent to use the tool by mentioning the tool name and describing what the function does.

For Claude Desktop

To add this MCP server to Claude Desktop:

1. Find your configuration file:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json
  • Linux: ~/.config/Claude/claude_desktop_config.json

2. Add this to your configuration file:

{
    "mcpServers": {
        "unity-mcp": {
            "command": "dotnet",
            "args": [
                "run",
                "--project",
                "path/to/unity-mcp"
            ]
        }
    }
}

3. Restart Claude Desktop for the changes to take effect

Want to 10x your AI skills?

Get a free account and learn to code + market your apps using AI (with or without vibes!).

Nah, maybe later