# IvanMurzak/Unity-MCP

AI Skills, MCP Tools, and CLI for Unity Engine. Full AI develop and test loop. Use cli for quick setup. Efficient token usage, advanced tools. Any C# method may be turned into a tool by a single line. Works with Claude Code, Gemini, Copilot, Cursor and any other absolutely for free.

Repository: https://github.com/IvanMurzak/Unity-MCP
Canonical: https://ross.abutalabs.com/products/ivanmurzak-unity-mcp
Homepage: https://ai-game.dev
Language: C#
License: Apache-2.0
License Family: permissive
Topics: ai, ai-integration, game-development, mcp, mcp-server, unity, claude, claude-code, claude-desktop, copilot, cursor, gemini, llm, model-context-protocol, openai, windsurf, gamedev, anthropic, cli
Last push: 2026-08-24T23:55:08+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 37
- inputs: {"age_days": 518, "days_push": 9, "days_rel": 9, "gap_med": 1, "n_releases_24m": 188}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3989, forks 363 (observed 2026-08-28T04:08:31.760487+00:00)

## What it is
A Unity plugin, MCP server, and CLI that lets AI agents (Claude, Cursor, Copilot, Gemini, Codex, etc.) build, edit, and playtest Unity games directly from the editor. It exposes 190+ MCP tools for scenes, GameObjects, prefabs, scripting, profiling, and domain extensions, with any C# method convertible to a tool in one line.

## Use cases
- let claude or cursor build and edit unity scenes via natural language
- connect an ai agent to the unity editor through mcp
- automate unity playtesting and iteration with an llm
- expose custom c# methods as mcp tools for ai agents
- reduce ai token usage when driving unity from an agent
- generate terrain, tilemaps, cinemachine cameras, or splines with ai
- run an ai develop-and-test loop inside unity

## When to choose
- you develop with Unity and want an AI agent to manipulate scenes, prefabs, components, or run scripts
- you use Claude Code, Cursor, Copilot, Gemini, Codex, or any MCP client and want it wired into Unity
- you want to expose your own C# methods as AI-callable tools with minimal code
- you need efficient token usage and local-first operation while iterating on a game
- you want domain-specific tool families like ProBuilder, Cinemachine, Terrain, Tilemap, Navigation, or Timeline

## When to avoid
- you work in Unreal or Godot and want the same experience without Unity-specific tooling
- you need a fully free hosted AI agent without bringing your own provider (paid tiers exist for hosted AI)
- you are not using an MCP-compatible AI client or editor
- you need offline operation with no internet access, since the plugin downloads its native MCP server binary on first run
- you require a stable, long-term-supported plugin for Unity versions older than 2022.3 LTS

## Facets
- artifact type: plugin
- maturity: active
- function: mcp, agent-framework, developer-tools, cli, workflow-automation, testing, game-engine
- domain: developer-tools, large-language-models
- platform: game-engine, cross-platform, windows, editor-plugin
- tags: unity, mcp-server, model-context-protocol, ai-agents, game-development, claude, cursor, copilot, gemini, codex, windsurf, openupm, csharp, unity-editor, playtesting, scene-manipulation, token-efficiency, local-first, probuilder, cinemachine, terrain, tilemap, navigation, timeline, godot, unreal-engine, automation, macos, linux, nodejs, docker

## Member repositories
- IvanMurzak/Unity-MCP (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:31.760487+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:24:13.497367+00:00, confidence not recorded.
  - readme: https://github.com/IvanMurzak/Unity-MCP (fetched 2026-08-28T04:08:31.760487+00:00, sha 5cd51b399cae)
  - homepage: https://ai-game.dev (fetched 2026-08-29T09:17:36.611736+00:00, sha 903b69660f51)
  - site_page: https://ai-game.dev/docs/quickstart (fetched 2026-08-29T09:17:36.623389+00:00, sha 0f7ed7e20464)
  - site_page: https://ai-game.dev/docs/app (fetched 2026-08-29T09:17:36.625719+00:00, sha bcf9b61e1e0b)
  - site_page: https://ai-game.dev/docs/auth (fetched 2026-08-29T09:17:36.628440+00:00, sha b2e756247881)
  - site_page: https://ai-game.dev/docs/faq (fetched 2026-08-29T09:17:36.630813+00:00, sha 790fa894c0d8)
  - site_page: https://ai-game.dev/docs/tools/unity (fetched 2026-08-29T09:17:36.632696+00:00, sha a500dcaabc7d)
  - site_page: https://ai-game.dev/docs/extensions/unity (fetched 2026-08-29T09:17:36.635002+00:00, sha d4aac788dfd8)
  - site_page: https://ai-game.dev/docs/unity-plugin-setup (fetched 2026-08-29T09:17:36.636539+00:00, sha 8c4cac937443)
  - site_page: https://ai-game.dev/pricing (fetched 2026-08-29T09:17:36.621163+00:00, sha b33eb057b256)
- Data as of 2026-08-30T08:39:29.467469+00:00.