Godot Sentinel

Provides an MCP server that connects AI assistants to a Godot project for real-time error detection, context-aware fixes, and git-backed patching.

typescript

0

GitHub Stars

typescript

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

Godot Sentinel MCP is an AI-powered development companion that connects your Godot project to MCP-compatible AI assistants. It watches your project in real time, surfaces errors, and automatically applies safe patches via git-backed workflows, so you can debug faster and keep your codebase healthy.

How to use

To use Sentinel MCP with an MCP client, first ensure the MCP server is running locally and your project is prepared. Start the server, connect your AI assistant, and interact with it using natural language queries. You can trigger automatic fixes, inspect project structure, run tests, and patch code directly from the AI assistant. The workflow emphasizes safety: every fix creates a new git branch, and you can easily rollback changes using standard git commands.

How to install

Prerequisites you need before installing Sentinel MCP are Node.js 18 or newer, Godot 4.3 or newer, Git, and gdUnit4 for testing.

Step by step installation and setup follow these commands:

npm install -g godot-sentinel-mcp

cd your-godot-project
sentinel init
# Creates a local environment configuration for Sentinel MCP

sentinel server
# Runs the MCP server locally on http://localhost:8787

# Connect your MCP client (for example Claude Code or another compatible MCP client) and start using AI-assisted fixes and project understanding

Usage patterns with an MCP client

Start the server and connect your chosen MCP client to gain access to the project context and real-time error handling.
Run tests through Sentinel MCP to validate fixes and see failures summarized by the AI assistant.
Use the AI assistant to understand project structure, inspect files, and apply safe patches that are automatically committed to a dedicated branch.

Available tools

run_tests()

Execute Godot tests and surface failures to the AI assistant.

get_context(file, line)

Retrieve code context around a specific location to inform AI fixes.

read_file(path)

Read the contents of a project file for inspection by the AI.

write_file(path, content)

Edit a file safely with AI-generated content.

apply_patch(diff)

Apply a git patch produced by the AI to modify code.

project_map()

Provide an overview of the project structure and key components.

list_movesets()

Show available movesets or data-driven content used by the game.

run_game(scene?)

Launch the game or a specific scene with monitoring enabled.