godot-mcp
MCP bridge between an AI assistant and the Godot editor, enabling control via 6 tools (call methods, run GDScript, screenshots, etc.) over WebSocket.
README
<a id="fr"></a>godot-mcp
Pont MCP entre un assistant IA et l'éditeur Godot.
Assistant IA <---stdio/MCP---> godot-mcp <---WebSocket:6505---> Plugin Godot
6 outils MCP pour piloter l'éditeur : godot_call (175+ méthodes), godot_list_methods, godot_info, godot_screenshot, godot_execute, godot_status.
Installation
1. Plugin Godot
Copiez le dossier plugin/ dans addons/godot_mcp/ de votre projet Godot.
votre-projet/
└── addons/
└── godot_mcp/ ← copiez plugin/ ici
Activez-le : Projet → Paramètres du projet → Plugins → godot-mcp → Activer
2. Serveur
npm install && npm run build && npm start
3. Client MCP
Ajoutez à la config de votre client IA (.mcp.json, claude.json, opencode.json) :
{
"mcpServers": {
"godot-mcp": {
"command": "node",
"args": ["/chemin/vers/godot-mcp/dist/index.js"]
}
}
}
Outils
| Outil | Description |
|---|---|
godot_call |
Appelle n'importe quelle méthode (175+) |
godot_list_methods |
Liste les méthodes par catégorie |
godot_info |
Infos projet |
godot_screenshot |
Capture éditeur en PNG |
godot_execute |
Exécute du GDScript |
godot_status |
Vérifie la connexion |
Chemins de nodes : les paramètres
parent_pathetnode_pathdegodot_call(ex.add_node,update_property,delete_node) sont toujours relatifs à la racine de la scène actuellement éditée ("."= racine de la scène). Les chemins absolus Godot ("/root","../...") sont rejetés — ils cibleraient l'arbre interne de l'éditeur au lieu de la scène.
godot_screenshotnécessite un éditeur avec rendu actif : il ne fonctionne pas en mode--headless(erreur "Could not get image from viewport").
Arborescence
godot-mcp/
├── plugin/ ← Plugin Godot (à copier dans votre projet)
├── src/index.ts ← Serveur MCP (Node.js)
├── package.json
├── tsconfig.json
├── README.md
└── LICENSE ← MIT
Licence
MIT
Ce plugin Godot dérive de godot-mcp-pro (Youichi Uda, MIT).
<a id="en"></a>godot-mcp
MCP bridge between an AI assistant and the Godot editor.
AI Assistant <---stdio/MCP---> godot-mcp <---WebSocket:6505---> Godot Plugin
6 MCP tools to control the editor: godot_call (175+ methods), godot_list_methods, godot_info, godot_screenshot, godot_execute, godot_status.
Setup
1. Godot Plugin
Copy the plugin/ folder into your project's addons/godot_mcp/.
your-project/
└── addons/
└── godot_mcp/ ← copy plugin/ here
Enable it: Project → Project Settings → Plugins → godot-mcp → Enable
2. Server
npm install && npm run build && npm start
3. MCP Client
Add to your AI client config (.mcp.json, claude.json, opencode.json):
{
"mcpServers": {
"godot-mcp": {
"command": "node",
"args": ["/path/to/godot-mcp/dist/index.js"]
}
}
}
Tools
| Tool | Description |
|---|---|
godot_call |
Call any method (175+) |
godot_list_methods |
List methods by category |
godot_info |
Project info |
godot_screenshot |
Editor screenshot in PNG |
godot_execute |
Run GDScript |
godot_status |
Check connection |
Node paths:
parent_pathandnode_pathparameters ofgodot_call(e.g.add_node,update_property,delete_node) are always relative to the root of the currently edited scene ("."= scene root). Absolute Godot paths ("/root","../...") are rejected — they would target the editor's internal tree instead of the scene.
godot_screenshotrequires an editor with active rendering: it does not work in--headlessmode ("Could not get image from viewport" error).
Structure
godot-mcp/
├── plugin/ ← Godot plugin (copy to your project)
├── src/index.ts ← MCP server (Node.js)
├── package.json
├── tsconfig.json
├── README.md
└── LICENSE ← MIT
License
MIT
This Godot plugin is derived from godot-mcp-pro (Youichi Uda, MIT).
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.