Unreal Engine MCP Bridge
A server that enables control of Unreal Engine 5.5+ using natural language commands, allowing users to create Blueprints, manipulate actors, work with nodes, manage project settings, and create UMG widgets through the Model Context Protocol.
README
Unreal Engine MCP (Model Context Protocol) Bridge
This project provides a bridge between text-based AI models and Unreal Engine 5.5+ using the Model Context Protocol (MCP). It allows you to control Unreal Engine using natural language commands and Python scripts.
Overview
The Unreal MCP Bridge enables:
- Creating and manipulating Blueprints
- Spawning and modifying actors in the level
- Working with Blueprint nodes and connections
- Managing project settings and input mappings
- Creating and modifying UMG widgets
- And much more!
Requirements
- Unreal Engine 5.5 or later with Python support enabled
- Python 3.10 or later
uvpackage manager (recommended) or pip
Installation
-
Clone this repository to your local machine
-
Install dependencies using
uv:
uv pip install -e .
Or using pip:
pip install -e .
- Make sure Unreal Engine is running with Python support enabled
Usage
Starting the MCP Server
To start the MCP server, run:
python unreal_mcp_server.py
This will start the server and register all available tools for interacting with Unreal Engine.
Using Example Scripts
The examples directory contains several example scripts that demonstrate how to use the MCP to interact with Unreal Engine:
add_cube.py: Spawns a cube actor in the level with customizable propertiescreate_rotating_cube_blueprint.py: Creates a Blueprint with a cube that rotates continuouslycreate_simple_game.py: Creates a simple game with a player character, collectibles, and score display
To run an example script:
python examples/add_cube.py
Available Tool Categories
The MCP Bridge provides tools in several categories:
Editor Tools
Tools for working with the Unreal Editor and actors in the level:
- Getting and finding actors
- Spawning and deleting actors
- Setting actor transforms and properties
- Taking screenshots
- Selecting actors
Blueprint Tools
Tools for creating and modifying Blueprint assets:
- Creating new Blueprints
- Adding and removing components
- Setting component properties
- Compiling Blueprints
Node Tools
Tools for working with Blueprint graph nodes and connections:
- Adding various node types (events, functions, variables)
- Connecting nodes
- Finding and deleting nodes
Project Tools
Tools for project-level operations:
- Creating input mappings
- Setting project settings
- Creating and managing levels
- Setting up GameModes
UMG Tools
Tools for creating and modifying UMG widgets:
- Creating Widget Blueprints
- Adding various widget types (Text, Button, Image, etc.)
- Binding widget events
- Managing widget properties
Extending the MCP
You can extend the MCP by adding new tools to the appropriate tool modules in the tools directory. Each tool should be registered with the MCP server in unreal_mcp_server.py.
Troubleshooting
- Make sure Unreal Engine is running before starting the MCP server
- Check that Python support is enabled in your Unreal Engine project
- Verify that the MCP server is running and listening for commands
- Look for error messages in the console output
License
This project is licensed under the MIT License - see the LICENSE file for details.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。