advanced-unity

Advanced Unity MCP allows you to control a Unity Editor session using natural language commands issued to an AI assistant. Through the Code Maestro MCP dashboard or compatible clients, you can request actions such as creating materials, managing assets and scenes, adjusting project and build settings, and running editor operations like play/pause, or triggering builds for specific platforms. The system exposes core editor controls, asset and scene management capabilities, build and platform tools, project configuration options, and basic performance analysis hooks, enabling a unified workflow between AI prompts and Unity projects. Supported clients include the Code Maestro Desktop App (direct connection) and other MCP clients like GitHub Copilot, Cursor, Windsurf, Claude Desktop, and more; the actual bridge for non-direct clients uses a relay server automatically installed by the MCP setup. To begin, install the package in Unity, connect via the MCP Dashboard, and start issuing commands such as creating a new material, switching the build target, or adding components to objects. The integration is designed to respond to natural language prompts by translating them into Unity Editor actions in real time.

Prerequisites:

• Unity 2022+ installed on your machine
• Internet access to fetch MCP components
• Code Maestro MCP Dashboard (or compatible MCP client) for connection management

Installation steps:

1. In Unity, open Package Manager > Add package from git URL and use the Unity-specific URLs from the README (for Unity 2020-2022 or Unity 6 paths) to install the Advanced Unity MCP package:
   • For Unity 2020-2022: https://github.com/codemaestroai/advanced-unity-mcp.git?path=Unity2020_2022
   • For Unity 6+: https://github.com/codemaestroai/advanced-unity-mcp.git?path=Unity6
2. After the package is added, open Code Maestro > MCP Dashboard in Unity to configure the connection mode.
3. If you are using Code Maestro Desktop App, select Direct Connection. If using other clients (VS Code, Cursor, etc.), choose Other Clients and click Configure for your client.
4. Ensure the Relay Server (for non-direct clients) is installed automatically by the setup; start Unity and the MCP dashboard to establish the connection.
5. Launch the Code Maestro Desktop App or your chosen client and begin issuing natural language commands to control the Unity Editor.

Notes:

• Depending on your workflow, you may also connect through a relay server for remote clients. The setup via the MCP Dashboard handles these details automatically.

Tips and common issues:

• Unity 2022+ is recommended for best compatibility with the Advanced Unity MCP features.
• Ensure the MCP Dashboard is configured to the correct connection mode before issuing commands.
• If multiple Unity instances are open, the AI will connect to the most recently opened Editor.
• Some actions require additional permissions or assets (e.g., creating new materials or modifying project settings) – ensure your project allows write access.
• When using the Relay Server path, ensure the local TCP bridge is reachable by your AI client and that the relay service is running.
• Review console messages in Unity if the AI reports errors; typical issues relate to object selection context or missing assets.
• You can issue compound commands like creating a scene with specific objects, or building the project for a target platform by phrasing natural language requests clearly.