godot-ai-mcp
Provides AI assistants with tools to launch the Godot editor, run projects, manipulate scenes, manage scripts, and control node properties through a standardized MCP interface.
README
Godot AI MCP Server
An enhanced Model Context Protocol (MCP) server for the Godot game engine. Provides AI assistants with tools to launch the editor, run projects, manipulate scenes, manage scripts, and control node properties — all through a standardized MCP interface.
Based on
@coding-solo/godot-mcpby Solomon Elias. That project provides the original core tools (launch editor, run project, scene management, UID tools). This fork adds script manipulation (set_script,read_script,write_script,list_scripts), scene inspection (get_scene_tree), node removal (remove_node), and property editing (set_node_property).
Original project: Coding-Solo/godot-mcp — MIT License.
Features
Core Tools
- launch_editor — Open the Godot editor for a project
- run_project — Run a Godot project in debug mode
- get_debug_output — Capture stdout/stderr from running projects
- stop_project — Stop a running project
- get_godot_version — Check installed Godot version
- list_projects — Find Godot projects in a directory
- get_project_info — Get project metadata (scene count, script count, etc.)
Scene Management
- create_scene — Create new scene files with any root node type
- add_node — Add nodes to existing scenes with properties
- remove_node — Remove nodes from scenes
- get_scene_tree — Inspect the full scene tree structure
- load_sprite — Load textures into Sprite2D/Sprite3D/TextureRect nodes
- export_mesh_library — Export 3D scenes as MeshLibrary for GridMap
- save_scene — Save/repack scene files
Script Tools
- set_script — Attach a script to a node (creates the script if it doesn't exist)
- read_script — Read
.gd,.cs, or.shaderfile contents - write_script — Create or overwrite script files
- list_scripts — List all scripts in a project
Node Properties
- set_node_property — Set any property on a node in a scene
UID Management (Godot 4.4+)
- get_uid — Get UID for a specific file
- update_project_uids — Resave all resources to update UID references
Requirements
- Node.js >= 18.0.0
- Godot 4.x (Mono build recommended for C# support)
Installation
npm install -g /path/to/godot-ai-mcp
Or use directly via npx:
npx /path/to/godot-ai-mcp
Configuration
MCP Config (opencode.json, claude desktop, etc.)
{
"mcp": {
"godot-ai": {
"type": "local",
"command": [
"env",
"GODOT_PATH=/path/to/your/godot/executable",
"DEBUG=true",
"npx",
"-y",
"godot-ai-mcp"
]
}
}
}
Environment Variables
| Variable | Description |
|---|---|
GODOT_PATH |
Path to the Godot executable (auto-detected if not set) |
DEBUG |
Set to true for verbose debug output on stderr |
Usage Examples
Create a scene with a Player node and attach a script
1. create_scene(projectPath="/path/to/project", scenePath="scenes/player.tscn", rootNodeType="CharacterBody2D")
2. add_node(projectPath="/path/to/project", scenePath="scenes/player.tscn", parentNodePath="root", nodeType="Sprite2D", nodeName="Sprite")
3. set_script(projectPath="/path/to/project", scenePath="scenes/player.tscn", nodePath="root", scriptPath="scripts/player.gd")
4. load_sprite(projectPath="/path/to/project", scenePath="scenes/player.tscn", nodePath="root/Sprite", texturePath="assets/player.png")
Inspect and modify a scene
1. get_scene_tree(projectPath="/path/to/project", scenePath="scenes/level.tscn")
2. set_node_property(projectPath="/path/to/project", scenePath="scenes/level.tscn", nodePath="root/Enemy", propertyName="position", propertyValue=[100, 200])
3. remove_node(projectPath="/path/to/project", scenePath="scenes/level.tscn", nodePath="root/DebugNode")
Script workflow
1. list_scripts(projectPath="/path/to/project")
2. read_script(projectPath="/path/to/project", scriptPath="scripts/enemy.gd")
3. write_script(projectPath="/path/to/project", scriptPath="scripts/enemy.gd", content="extends CharacterBody2D\n...")
Building from Source
npm install
npm run build
The compiled output goes to build/. The GDScript file is copied to build/scripts/.
License
MIT — see LICENSE for details.
This project is derived from @coding-solo/godot-mcp (MIT, © Solomon Elias). The original core server architecture, GDScript headless operations pattern, and scene manipulation tools are based on that project.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.