This repository provides a seamless integration between Model Context Protocol (MCP) and Unity, enabling AI assistants to understand and interact with Unity projects in real-time. AI can access scene data,
This package provides a seamless integration between the Model Context Protocol (MCP) and the Unity Editor, enabling AI assistants to interact with Unity projects in real-time. It allows AI to access scene hierarchy, project settings, and execute code directly within the Unity Editor context.
Key Features: Project file manipulation, real-time project information access, scene hierarchy understanding, C# code execution, log monitoring, and editor play mode control.
Getting Started: Requires Unity 2021.3+ and Node.js 18+. Installation involves adding the package via Git URL or importing a custom package. The MCP server can be run directly using Node.js or added to an MCP Host configuration.
Usage: The MCP Debug window allows connection monitoring and feature testing. Available tools include Unity Editor tools (e.g., get_editor_state
, execute_editor_command
) and filesystem tools (e.g., read_file
, edit_file
).
Architecture: Consists of a Unity plugin (C#) and an MCP server (TypeScript/Node.js) communicating via WebSocket.
License: MIT License.
quazaai/UnityMCPIntegration
March 17, 2025
March 28, 2025
C#