Unreal

A MCP Server implementation for interacting with Unreal Engine instances through remote Python execution.

html

5

GitHub Stars

html

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": {
    "radial-hks-mcp-unreal-server": {
      "command": "python",
      "args": [
        "-m",
        "src.mcp_server_unreal.server"
      ]
    }
  }
}

You can manage Unreal Engine instances remotely by running Python-based MCP Unreal Server, which discovers Unreal nodes, executes Python code in those environments, and provides detailed logging and health monitoring. This server makes it straightforward to control Unreal instances, run scripts, and evaluate expressions from a centralized MCP client.

How to use

You will connect your MCP client to the Unreal MCP server to discover available Unreal nodes, monitor their status, and execute Python code inside Unreal Engine environments. Use the client to send requests to run Python files, evaluate expressions, or execute statements in attended or unattended modes. You can perform file execution or inline code execution, and you will see node health and resource information reported back in real time.

How to install

# Prerequisites
Python 3.8+ is required
pip is available in your Python environment

# Clone repository
git clone https://github.com/your-org/mcp-unreal-server.git
cd mcp-unreal-server

# Install dependencies
pip install -r requirements.txt

# Run the server (stdio command)
python -m src.mcp_server_unreal.server

Configuration and notes

Configure multicast-based remote execution and logging to tailor how the server discovers Unreal nodes and logs activity. In code, set the multicast group endpoint to your network’s multicast address and port. You can also adjust logging verbosity to control what information is written to files and shown on the console. The server writes detailed logs to a file named mcp_unreal.log and prints important information to the console so you can monitor node connections and status in real time.

Troubleshooting

Common issues include no nodes being discovered, execution timeouts, or connection drops. Ensure Unreal instances are running with the MCP plugin enabled for discovery, verify multicast traffic is allowed by your firewall, and review mcp_unreal.log for node status changes and error messages.

Notes on usage patterns

The server supports both attended and unattended execution modes, as well as file execution and statement evaluation. When you start the server, you can connect your MCP client to perform actions such as listing available Unreal nodes, executing Python code within an Unreal Engine environment, or running Python files across multiple Unreal nodes for batch automation.