mcpField
G
godot-mcp-bridge

godot-mcp-bridge

Local
github.com/TomasLucasUTNDeveloper Tools10Updated 8w ago
Repository

About godot-mcp-bridge

godot-mcp-bridge is an MCP server that lets AI assistants work directly with a running Godot editor. It covers scene and script editing, project settings, and project visualization, with open-scene edits using Godot’s live tree and undo system.

Key features

  1. 01225 tools across editor tasks
  2. 02Edit open scenes through the live tree with undo
  3. 03Browse script relationships in a project visualizer
  4. 04Edit code and scene node properties in the visualizer

Use cases

  1. 01Create or modify scenes and scripts from an AI client
  2. 02Inspect script connections before refactoring
  3. 03Explore project structure through a graph view

MCP support

It exposes Godot editor operations as MCP tools for compatible AI clients. The server runs locally and supports one Godot instance at a time.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "tomaslucasutn-godot-mcp-bridge": {
      "command": "npx",
      "args": [
        "-y",
        "godot-mcp-bridge"
      ]
    }
  }
}
Run commandnpx -y godot-mcp-bridge

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmgodot-mcp-bridge1.2.1

Tools

44

Saved tool list · updated 2w ago

This tool takes no parameters.

Details

From the project README

godot-mcp-bridge

Give your AI assistant full access to the Godot editor.

Build games faster with Claude, Cursor, or any MCP-compatible AI — no copy-pasting, no context switching. The AI reads, writes, and manipulates your scenes, scripts, nodes, and project settings directly inside the running Godot editor.

Godot 4.5+ · 225 tools (35 loaded by default) · Live-tree editing with undo · Interactive project visualizer · MIT license

See the project README for the full tool breakdown, architecture, and setup guide. This file is the condensed version shown on npm.

Quick Start

1. Install the Godot plugin
npx godot-mcp-bridge install

Run from inside your Godot project folder — copies the addon into addons/godot_mcp/ and enables the plugin. Or copy addons/godot_mcp/ from the GitHub repo by hand and enable it under Project Settings → Plugins. (The "Godot AI Assistant tools MCP" AssetLib listing belongs to the upstream project this forked from, not this repo.)

2. Add the server to your AI client

Claude Desktop / Cursor — add to the client's MCP config:

{
  "mcpServers": {
    "godot": {
      "command": "npx",
      "args": ["-y", "godot-mcp-bridge"]
    }
  }
}

Windows: use "command": "cmd", "args": ["/c", "npx", "-y", "godot-mcp-bridge"].

Works with any MCP-compatible client (Claude Code, Cline, Windsurf, etc.)

3. Connect

Restart your AI client, then restart your Godot project. Check the top-right corner of the editor for MCP Connected in green.

What Can It Do?

225 tools grouped by intent; only core (35 tools) loads by default so the agent stays focused. enable_toolset({ name: "..." }) pulls in more — runtime control, scene editing, project config, animation, physics, analysis, scaffolding, testing, and more. Edits to a scene that's open in the editor go through the live tree and Godot's undo system instead of silently overwriting the .tscn file on disk.

Interactive Visualizer

Run map_project and get a browser-based project explorer at localhost:6510:

  • Force-directed graph of all scripts and their relationships
  • Click any script to see variables, functions, signals, and connections
  • Edit code directly in the visualizer — changes sync to Godot in real time
  • Scene view with node property editing
  • Find usages before refactoring

Current Limitations

  • Local only — runs on localhost, no remote connections
  • Single connection — one Godot instance at a time
  • AI has limited Godot knowledge — it can help debug, write scripts, and build scenes, but can't create a complete game without guidance

Build From Source

git clone https://github.com/TomasLucasUTN/godot-mcp-bridge
cd godot-mcp-bridge/mcp-server
npm install
npm run build

Then point your AI client at mcp-server/dist/index.js instead of using npx.

License

MIT

GitHub · Report Issues

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.