isuzu-shiranui/UnityMCP

Drive the Unity Editor from an AI agent or the terminal. The Editor serves MCP itself over HTTP, so there is no second process to run, and the isuzu-unity-cli command needs no Node or .NET runtime.

What it solves

UnityMCP opens the Unity Editor to AI agents, allowing them to interact with the editor's internal state, assets, and scene hierarchy. It eliminates the need for manual intervention by providing a standardized way for AI agents to execute editor commands, diagnose project issues, and automate authoring tasks.

How it works

The framework turns the Unity Editor into an MCP (Model Context Protocol) server by exposing a streamable HTTP endpoint on the local machine. AI agents (such as Claude Code, Cursor, or Gemini CLI) connect directly to this endpoint to call tools. Developers can define new tools simply by adding the [McpTool] attribute to static C# methods, which automatically generates the necessary JSON Schema for the agent to understand the tool's function.

Who it’s for

Unity developers who want to integrate AI agents into their development workflow to automate repetitive editor tasks, perform project diagnostics, or programmatically manipulate game objects and assets.

Highlights

  • Direct Integration: Connects AI agents directly to the Unity Editor without requiring a separate external MCP server process.
  • Extensible Tooling: Allows easy creation of custom tools using C# attributes or JSON definitions.
  • Comprehensive Toolset: Includes built-in tools for diagnostics (console logs, compilation status), authoring (creating GameObjects, Prefabs, and Animator Controllers), and input synthesis.
  • Cross-Client Support: Compatible with major AI clients including Claude Code, Claude Desktop, Cursor, Codex, and Gemini CLI.
  • CLI Access: Includes a native isuzu-unity-cli for executing tools from the command line without needing Node or .NET runtimes.

Related

  • Project
    csml-rpi/Foam-Agent

    An end-to-end multi-agent framework that automates OpenFOAM CFD simulations from natural language prompts, featuring automatic error correction and RAG-enhanced configuration generation.

  • Project
    agentclientprotocol/python-sdk

    A Python SDK for the Agent Client Protocol (ACP) that provides typed schema models and asyncio transports for building standardized AI agents and clients.

  • Project
    kris-hansen/comanda

    A terminal-native runtime for durable, self-improving agentic workflows in a repository, using quality gates and codebase indexing to ensure work is actually completed.

  • Project
    lingo-db/lingo-db

    LingoDB is a high-performance data processing system that uses compiler technology and declarative sub-operators to support flexible, cross-domain workflows beyond traditional SQL.