db-lyon/ue-mcp

Complete Unreal Engine development toolkit exposed as MCP tools.

UE‑MCP – Unreal Engine Model‑Context‑Protocol Server

What it is – A Node.js/TypeScript server that lets an AI assistant talk to the Unreal Editor via a C++ bridge plugin. It exposes ~2 040 editor actions (levels, blueprints, materials, VFX, PCG, etc.) as JSON‑RPC calls, so natural‑language commands can create, read, or modify a UE 5 project.

How it works

  • ue-mcp installs a small C++ plugin into your UE project and starts a WebSocket‑JSON‑RPC server.
  • The AI client sends plain‑English (or structured) commands; the server translates them into UE API calls through the bridge.
  • Actions are also composable in a ue-mcp.yml YAML flow, enabling multi‑step, repeatable workflows.
  • An npm‑based plugin system lets developers add new actions or whole categories.

Key features

  • 29 tool categories covering everything from level layout to animation, Niagara, PCG, GAS, networking, and the new UE 5.8 native toolset.
  • Built‑in 19‑step demo that procedurally builds a “Neon Shrine” scene entirely via the bridge.
  • Safety guards – write‑guards, source‑control checks, and flow rollback to protect projects.
  • Extensible – publish npm plugins (ue-mcp‑<plugin>) that inject new actions.
  • YAML flows – define repeatable pipelines with retries, rollbacks, and custom JS/TS tasks.
  • Cross‑platform – works on Windows, macOS, Linux (UE 5.4‑5.8).

Typical use cases

  • Let a large‑language‑model (e.g., Claude, Cursor) iteratively design levels, assets, or gameplay systems.
  • Automate repetitive editor tasks (batch import, material generation, build pipelines).
  • Prototype procedural content generation pipelines without manual UI work.
  • Provide AI‑driven “assistant” features inside a game‑dev studio’s tooling.

Getting started

npx ue-mcp init   # detects .uproject, deploys bridge, enables required UE plugins
npx ue-mcp doctor # troubleshoot connection issues

After the editor restarts, send commands like:

"Place a point light 300 units above the player start, intensity 5000."
"Build the project in Development and report any compile errors."

Installation requirements

  • Unreal Engine 5.4‑5.8 (Windows) or 5.6+ (Linux/macOS).
  • Node 18+.
  • A C++ toolchain (Visual Studio on Windows, Xcode CLI on macOS, gcc/clang on Linux).
  • PythonScriptPlugin (bundled with UE) – enabled automatically by init.

Documentation & community

  • Full docs: https://ue-mcp.com/docs (getting‑started, tool reference, flows, plugins, troubleshooting).
  • MIT‑licensed, sponsor‑friendly, accepts PRs and issues.

Limitations / considerations

  • Requires a compiled C++ bridge; first‑run may fail if the editor isn’t restarted or the toolchain is missing.
  • Write access gives the AI full control over project files – safety guards are optional but recommended.
  • Only supports UE 5; older engine versions are not covered.

License – MIT (free for personal, educational, and commercial use).

Related

  • Project
    ChiR24/Unreal_mcp

    EngineWire is a native Unreal Engine plugin that lets LLM‑based assistants (Claude, Cursor, VS Code, etc.) control the editor through a single MCP tool called `unreal`. It exposes ~400 actions—spawning actors, editing Blueprints, building materials, running Play‑In‑Editor, taking screenshots, and more—via a simple search/describe/execute flow over either a local HTTP server (port 3000) or a Node.js stdio proxy (port 8090). Security is token‑based and localhost‑only by default. Install the plugin, enable the server, add the token to your AI client, and you can issue natural‑language commands that the plugin translates into Unreal‑Editor operations.

  • Project
    isuzu-shiranui/UnityMCP

    A framework that exposes the Unity Editor as an MCP server, enabling AI agents to automate editor tasks, diagnose projects, and manipulate assets.

  • Project
    AnkleBreaker-Studio/unity-mcp-server

    AnkleBreaker Unity MCP is a Model Context Protocol server that lets AI assistants control multiple Unity editors simultaneously — building scenes, running tests, managing multiplayer Play Mode, and inspecting performance — with per-editor queues, action history, and undo support. It's built for teams coordinating several agents across projects and demanding Unity workflows.

  • Project
    CoderGamester/mcp-unity

    An implementation of the Model Context Protocol for the Unity Editor, allowing AI assistants to directly manipulate scenes, assets, and editor settings.

  • Project
    IvanMurzak/Unity-MCP

    Unity‑MCP is a Unity package that connects LLM agents (Claude, Codex, etc.) to the Unity Editor and to compiled games. It provides 70+ built‑in “tools” (asset management, scene manipulation, scripting, testing, profiling, etc.) that the AI can call, enabling natural‑language creation, debugging, and runtime AI behaviour. Install via a UnityPackage, OpenUPM, or the `unity-mcp-cli`; configure an agent, auto‑generate a skill set, and start issuing commands like “create three cubes in a circle”. Extensible with optional skill packs for Animation, Cinemachine, Navigation, and more. MIT‑licensed.