What are the system requirements?
Unreal Engine 5.5+ and Python 3.12+ are required. An MCP-compatible client (Claude Desktop, Cursor, or Windsurf) is also needed.
@chongdashu
基本信息
Enable AI assistant clients like Cursor, Windsurf and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP).
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"unrealMCP": {
"command": "uv",
"args": [
"--directory",
"<path/to/the/folder/PYTHON>",
"run",
"unreal_mcp_server.py"
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
An experimental integration that enables AI assistant clients like Cursor, Windsurf, and Claude Desktop to control Unreal Engine through natural language using the Model Context Protocol (MCP). It consists of a UE plugin (C++ TCP server) and a Python MCP server that expose tools for actor, Blueprint, and editor operations.
You use the provided Unreal sample project or copy the plugin into your own project, enable the plugin, and build it. Then set up the Python server (requires Python 3.12+), and configure your MCP client (e.g., Claude Desktop, Cursor, Windsurf) with a JSON configuration pointing to unreal_mcp_server.py. The server communicates with the plugin over TCP (port 55557).
Unreal Engine 5.5+ and Python 3.12+ are required. An MCP-compatible client (Claude Desktop, Cursor, or Windsurf) is also needed.
Add a JSON entry under `mcpServers` in your client’s configuration file, specifying `uv` as the command with arguments pointing to the Python directory and `unreal_mcp_server.py`. The file location varies: Claude Desktop uses `~/.config/claude-desktop/mcp.json`, Cursor uses `.cursor/mcp.json` in the project root, and Windsurf uses `~/.config/windsurf/mcp.json`.
No, the project is currently experimental. Features may be incomplete or unstable, breaking changes can occur without notice, and production use is not recommended.
The Python server uses TCP sockets (port 55557) to communicate with the Unreal Engine C++ plugin, and implements the Model Context Protocol via the FastMCP library.
You can reach the author on X/Twitter: [@chongdashu](https://www.x.com/chongdashu).

Search a registry of agent skills and MCP servers, run them for real (sandbox or remote MCP, no install, a trace on every run), and hire proven agents for USDC.
Model Context Protocol (MCP) server for Hostinger API. If you don't want to install or run anything locally, connect directly to Hostinger's hosted MCP server: This opens a browser window to authorize via OAuth. Once app

ATL doesn’t just help agents find tools — it gets the job done. It finds and chooses the right AI provider, tool, or MCP server, executes the task, and switches to a fallback when a provider fails. It can optimize for lo
Detect and redact PII, PHI, PCI and secrets — SSNs, credit cards, passports, API keys and cloud credentials — in text and in local files, using the Strac DLP API. Five tools: redact_text, detect_sensitive_data, detect_fi
28-tool MCP server for web scraping, crawling, deep research & autonomous extraction — clean Markdown & structured JSON for Claude, Cursor & any MCP client. 1,000 free one-time credits, local-Ollama LLM support.
评论