Log inSubmit server
Unity MCP logo

Unity MCP

by IvanMurzak

Connects AI agents to Unity Editor and compiled games for automation, code generation, testing, debugging, and runtime AI interactions.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Unity MCP

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "ivanmurzak-unity-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "unity-mcp-cli"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

IvanMurzak/Unity-MCP connects MCP-compatible AI agents to both the Unity Editor and compiled Unity games. It provides built-in tools for operating on Unity projects, generating code, testing game mechanics, and debugging issues, while also supporting in-game LLM features such as dynamic NPC behavior. Use it when an agent needs to make changes inside a Unity project or interact with a running game. Connections can run locally over stdio or remotely over HTTP.

Use cases

  • Automate Unity project and asset operations
  • Generate and test game mechanics
  • Debug errors in Unity projects with an AI agent
  • Add dynamic NPC behavior to a compiled game
  • Create project-specific MCP tools

Key features

  • Unity Editor MCP integration
  • Compiled-game runtime support
  • 70+ built-in tools
  • Local stdio and remote HTTP deployment
  • Project-aware skill generation
  • Custom tool extensibility

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

28/28 checks healthy over the last 77d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 7d ago via OSV.dev · unity-mcp-cli (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars4.5kStargazers on the source repository.
npm downloads14kPackage downloads in the last 30 days.
Last commitTodayMost recent push to the default branch.
Directory activity1 upvotesConfig copies, upvotes, and views on AllMCPs.

How Unity MCP works

What IvanMurzak/Unity-MCP does

IvanMurzak/Unity-MCP exposes Unity development and runtime functionality through the Model Context Protocol. It is designed for AI-assisted game development in the Unity Editor, with an additional runtime mode that allows an LLM to operate inside a compiled game.

The package includes more than 70 built-in tools grouped into areas such as project and asset management. These tools let an agent carry out Unity workflows rather than only generate text or source code. The project also supports creating custom tools in project code, so Unity teams can add project-specific operations to the MCP interface.

Typical requests include creating scene objects, developing game mechanics, running tests, investigating errors, and asking an agent to modify a project through natural-language instructions. Runtime support extends these workflows to applications such as in-game debugging, player interaction, and dynamic NPC behavior.

How it works

The Unity package runs within the Unity project and exposes MCP capabilities to an external AI client. The client sends tool requests through a local stdio connection or a remotely configured HTTP connection. Agents from different providers can be used, including Anthropic, OpenAI, and Microsoft models; the integration is not tied to one model vendor.

Unity-specific skills can be generated from the project. The generated skills account for factors such as the operating system, Unity version, and installed project plugins, giving the connected agent project-relevant context. AI clients can then use the available tools to make changes, inspect project state, generate code, and help diagnose problems.

The runtime mode packages the connection into a compiled Unity game. This enables AI features to execute while the game is running instead of limiting MCP access to the Editor.

Setup and configuration

Install the Unity package with the downloadable .unitypackage installer or through OpenUPM:

bash
openupm add com.ivanmurzak.unity.mcp

An alternative command-line workflow installs the plugin into a project:

Terminal
npm install -g unity-mcp-cli
unity-mcp-cli install-plugin ./MyUnityProject

After opening the project, use Unity’s Window/AI Game Developer panel. The recommended setup is to select Auto-generate skills; the panel also provides a Configure MCP option for manual configuration.

The CLI can open the project and connect it automatically:

bash
unity-mcp-cli open ./MyUnityProject

For supported workflows, skills can also be configured from the command line, for example with unity-mcp-cli setup-skills claude-code ./MyUnityProject. The project supports local stdio deployment and remote HTTP deployment.

Tools and capabilities

  • Operate on Unity projects and assets through built-in MCP tools.
  • Generate and modify game code with an AI agent.
  • Develop and test game mechanics.
  • Investigate and fix project problems with AI-assisted debugging.
  • Use AI inside compiled Unity games.
  • Create custom MCP tools in project code.
  • Generate project-aware skills based on environment and Unity configuration.

Limitations and notes

The integration requires a Unity project with the package installed and an MCP-compatible AI client configured to connect to it. Editor workflows and compiled-game workflows are both supported, but runtime behavior requires the runtime connection to be included in the game. The available built-in tools are documented separately in the project’s default MCP tools reference.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Unity MCP

IvanMurzak/Unity-MCP is an MCP server that connects AI agents to the Unity Editor and compiled Unity games. Its main tools support Unity project and asset operations, code generation, testing, debugging, and runtime interactions such as in-game AI behavior. It also supports project-specific custom tools.

Install the Unity package with the downloadable `.unitypackage` installer or run `openupm add com.ivanmurzak.unity.mcp`. The `unity-mcp-cli` alternative can install the plugin with `unity-mcp-cli install-plugin ./MyUnityProject`.

Yes, IvanMurzak/Unity-MCP works with Cursor. The project also documents connections from Claude, Windsurf, Claude Code, GitHub Copilot, Codex, and Cline, with setup performed through the Unity `Window/AI Game Developer` panel or the CLI.

Yes, IvanMurzak/Unity-MCP supports runtime use inside compiled Unity games. This enables in-game LLM features, real-time debugging, player-AI interaction, and behaviors such as dynamic NPC responses.

Related MCP servers

All in Gaming All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewUnity MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/ivanmurzak-unity-mcp?style=directory)](https://allmcps.com/mcp/ivanmurzak-unity-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/ivanmurzak-unity-mcp"><img src="https://allmcps.com/api/badge/ivanmurzak-unity-mcp?style=directory" alt="Unity MCP on AllMCPs" /></a>