ue5 -bridge

This MCP server acts as a bridge between AI assistants and Unreal Engine 5, exposing a Model Context Protocol interface that lets you manipulate the editor through natural-language commands. It supports a wide range of tools for level and actor management, asset handling, Blueprint and Animation Blueprint operations, and script execution. Clients that implement MCP can connect to the HTTP endpoint exposed by this server and issue commands like spawning actors, listing level actors, opening levels, modifying properties, or creating and configuring Blueprints. The server is designed to plug into Unreal Editor workflows and accelerate common tasks with AI-assisted prompts.

To use it, run the server locally and configure your MCP client to point at the server’s endpoint (for example http://localhost:3000). In your client’s MCP configuration, specify the server as an MCP provider and reference the node-based bridge script (index.js) as the entrypoint. Once connected, you can issue natural-language commands such as spawning a light at a position, listing StaticMeshActors, creating a Blueprint, or adding variables to a Blueprint. The server exposes a comprehensive set of tools organized by capabilities (connection, level/actor control, assets, scripts, viewport, Blueprints, Animation Blueprint operations, characters, materials, and enhanced input).

Supported workflow examples include: spawning actors, adjusting actor properties, querying the current level’s actors, executing Python/C++/console scripts, and querying or modifying Animation Blueprints. The aim is to speed up UE5 development by handling repetitive editor tasks via natural language rather than manual clicking.

Prerequisites:

1. Node.js v18 or newer installed on your machine.
2. Git installed to clone the repository.
3. A running Unreal Editor instance (optional for local testing) and an HTTP endpoint the MCP server can forward to if you’re bridging to Unreal via HTTP.

Installation steps:

1. Clone the repository: git clone https://github.com/Natfii/ue5-mcp-bridge.git
2. Change into the project directory: cd ue5-mcp-bridge
3. Install dependencies: npm install
4. Run the MCP bridge (development): npm run start
5. Verify the server starts and listens on the configured port (default could be http://localhost:3000).

Configuration note:

• Ensure your MCP client points to the server’s URL and uses the correct entry script (index.js) as shown in the Quick Start configuration.

Tips and notes:

• The server is Node.js-based and expects an MCP-compatible client to connect. If you deploy behind a reverse proxy, ensure proper routing to the MCP endpoint.
• The Quick Start examples show how to configure popular MCP clients (Claude Code, Claude Desktop) to connect to this server via the node bridge entry point.
• Environment variables are optional; you can add them to the mcpServers[<server-name>].env object if your deployment requires configuration, e.g., API keys or URLs for connected services.
• If you encounter connection issues, verify that Unreal Editor is accessible from the same host (or adjust network permissions) and that the REST/API endpoint matches what the client expects.
• This bridge supports a wide array of tools (spawn_actor, get_level_actors, set_property, unreal_blueprint_modify, unreal_anim_blueprint_modify, unreal_execute_script, etc.). Refer to the Available Tools section in the README for the complete list and usage semantics.