OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
MCP & Tooling

MCP for Unity

MCP for Unity is an open-source C# integration that connects MCP-compatible AI assistants and language models to the Unity Editor. It exposes tools for asset management, scene and GameObject control, C# script editing, testing and workflow automation, plus read-only resources for inspecting editor and project state.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

MCP for Unity — OpenViST project profile in MCP & Tooling.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want an MCP-compatible assistant to make or inspect changes in a Unity Editor project, rather than only discuss Unity development outside the editor.

Who it is for

  • Unity developers who want an AI assistant to act on an open project in the Editor.
  • Teams or individual developers automating repetitive editor work, testing or asset processing.
  • Developers building custom AI-driven Unity editor tools on MCP.

Documented uses

  • Create or modify scenes and GameObjects from assistant prompts.
  • Manage and process project assets through MCP tools.
  • Generate, edit or refactor C# scripts with project context.
  • Run tests and automate validation or other repetitive editor workflows.
  • Build custom AI-driven editor tools using the MCP protocol.
Before you choose

Requirements & limits

The repository identifies the software as MIT-licensed. The documented setup requires Unity 2021.3 LTS or newer, Python 3.10+ with uv, and an MCP client.

  • It is an integration layer for the Unity Editor, so using it requires a Unity project and a separate MCP client.
  • The documented install offers Git URL, Unity Asset Store and OpenUPM paths.
  • Transport and extra setup vary by MCP client; Claude Desktop uses stdio, and several clients use HTTP by default.
  • The project documents multi-instance support for controlling several Unity Editors from a session.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install the Unity package and connect an MCP client

Set up the Unity package, Python server and a compatible MCP client using the documented Git URL path.

Before you start

  • Unity 2021.3 LTS or newer
  • Python 3.10+ with uv
  • A compatible MCP client
  • Git available on PATH for the Git URL installation path

Steps

  1. In Unity, open Window → Package Manager, select the + button, and choose Add package from git URL.

  2. Paste https://github.com/CoplayDev/unity-mcp.git?path=/MCPForUnity#main and import the package.

  3. Follow the setup wizard to confirm Python and uv, then choose detected clients and configure them.

  4. In the MCP client, try a documented prompt such as creating colored cubes in the current scene.

Expected outcome

The Unity package and server are configured for a selected client, which can send prompts to the Unity Editor.

Guide limitations

  • Client setup varies: Claude Desktop uses stdio, while HTTP is the default for several other clients.
  • Some clients require enabling an MCP toggle or plugin in their own settings after configuration.
  • The guide's Git URL installation path requires Git on PATH.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗Guide documentation ↗
Source record details
Source availability
Active
Sources checked
1 October 2026
Original website title
CoplayDev/unity-mcp · Official homepage title · retained in its source language
How we review sourcesSuggest a correction