AIAny
Icon for item

MCP for Unity

Bridges LLM-driven AI assistants to the Unity Editor so models can create scenes, edit C# scripts, manage assets, run tests and automate game-dev workflows. Exposes 47 focused MCP tool entrypoints, supports many MCP clients, and is MIT-licensed for local use.

Introduction

Most game teams waste time on repetitive Editor tasks that are trivial to express in language but expensive to script and maintain. Letting an LLM operate the Unity Editor via a standard protocol turns those micro-tasks into prompts you can delegate, inspect, and iterate on.

What Sets It Apart
  • Protocol-first integration: implements the Model Context Protocol (MCP) so any MCP-capable client (Claude Desktop/Code, Cursor, VS Code, Windsurf, Cline, Gemini CLI, etc.) can control Unity without custom per-client adapters — so what: you can swap assistants or run local models with minimal reconfiguration.
  • Editor-native toolset: exposes ~47 focused tool entrypoints (create GameObjects, edit scripts, run tests, profile, build, asset ops) rather than a single monolithic API — so what: prompts can call intent-specific tools with predictable effects and easier validation.
  • Lightweight local server model: runs alongside Unity (requires Python 3.10+ via uv) and auto-configures many clients — so what: avoids cloud-side build/deploy for day-to-day Editor automation and keeps assets/code local.
  • Open-source & permissive license: MIT license and visible contributor history — so what: teams can audit, modify, or integrate MCP for Unity into CI/tooling without restrictive licensing.
Who It's For and Tradeoffs

Great fit if you maintain Unity projects and want to speed authoring by delegating repetitive Editor tasks to LLMs, experiment with prompt-driven content generation, or integrate assistant-driven workflows into your pipeline. It also helps toolsmiths who want to build MCP-aware plugins that target multiple assistants. Look elsewhere if you need a hosted/cloud authoring service, require enterprise-grade multi-user locking out of the box, or prefer a no-dependencies Unity Package (MCP for Unity requires a small Python runtime and the uv tool for the server).

Where It Fits

MCP for Unity sits between assistant clients and the Editor: use it when you want your LLM or local model to directly manipulate scenes, scripts, and assets with clear, inspectable actions. For purely runtime AI integration (in-game agents, inference services), pair MCP for Unity with separate runtime SDKs rather than using it as an in-game AI runtime.

Information

  • Websitegithub.com
  • OrganizationsCoplayDev, Aura
  • Published date2025/03/18

More Items

GitHub

Connects MCP-compatible AI assistants to Google Ads for account queries and reporting, with guarded support for asset creation, video uploads, and label management. It does not modify campaigns, budgets, or bids.

GitHub

Inspect and measure 3D scans in the browser, compare them with CAD models, and reverse-engineer them into STEP solids. An optional MCP connection lets AI agents operate the workflow; without it, files stay local.

Hugging Face
AI Video2026

Remakes existing footage from text instructions, changing characters, props, or scenes while preserving the original motion, camera, cuts, and sound. Use it in the Viggle app or run it on your own GPUs.