GoPeak logo

GoPeak

Community
HaD0Yun

GoPeak — The most comprehensive MCP server for Godot Engine. 95+ tools: scene management, GDScript LSP, DAP debugger, screenshot capture, input injection, ClassDB introspection, CC0 asset library. npx gopeak

PublisherHaD0Yun
RepositoryGopeak-godot-mcp
LanguageJavaScript
Forks
17
Stars
189
Available tools
33
Transport typestdio
Categories
LicenseMIT
Links
  • Connect tools to AI workflows

    GoPeak exposes MCP capabilities that can be used by compatible AI clients and agents.

  • 33 available tools

    Browse the callable actions below, including names and descriptions when provided by the server.

  • Ready-to-copy setup

    Use the installation snippets to configure this server in your preferred MCP client.

  • Open source signals

    189 stars and 17 forks from the linked repository.

GoPeak

Made with Godot npm

🌐 Languages: English | 한국어 | 日本語 | Deutsch | Português | 简体中文

GoPeak Hero

GoPeak is an MCP server for Godot 4 that gives AI assistants a real edit → run → inspect → fix loop.

It is designed for trusted Godot 4 workflows: small default tool surface, setup-gated advanced capabilities, and explicit compatibility rules for older/legacy tool names.

English is the canonical source of truth. Localized READMEs are concise overviews and may lag behind README.md.

Discord is temporarily unavailable while the invite link is refreshed. Use GitHub Discussions for now.


Quick Start

Requirements

  • Godot 4.x
  • Node.js 18+
  • MCP-compatible client such as Claude Desktop, Cursor, Cline, or OpenCode

1) Run GoPeak

bash
npx -y gopeak

or install globally:

bash
npm install -g gopeak
gopeak

2) Add MCP client config

json
{
  "mcpServers": {
    "godot": {
      "command": "npx",
      "args": ["-y", "gopeak"],
      "env": {
        "GODOT_PATH": "/path/to/godot",
        "GOPEAK_TOOL_PROFILE": "compact"
      }
    }
  }
}

compact is the default profile. It keeps the initial MCP context small and exposes additional setup-gated groups only when requested.

3) Try these prompts

  • "List Godot projects in /your/projects and show project info."
  • "Create scenes/Player.tscn with a CharacterBody2D root and a movement script."
  • "Run the project, read the debug output, and fix the top error."
  • "Use tool.catalog to find animation tools, then activate the right group."

What You Get

WorkflowWhat GoPeak can do
Project controlFind projects, launch the editor, run/stop the game, collect debug output.
Scene + script editingCreate scenes, add nodes, edit typed properties, create/modify GDScript.
Resource workflowsWork with resources, materials, shaders, imports, and export-related checks.
DebuggingUse logs, Godot LSP diagnostics, DAP breakpoints/stack traces, and runtime inspection when configured.
Runtime testingCapture screenshots, inspect live trees, inject input, and call runtime methods through the addon.
Tool discoveryKeep the default surface compact, then activate capability groups with tool.catalog or tool.groups.

Setup gates

Some capabilities require extra Godot-side services. GoPeak labels these instead of pretending everything is always available:

CapabilityRequires
Editor bridge scene/resource editsgodot_mcp_editor plugin enabled in the Godot project.
Runtime inspection, screenshots, input injectionRuntime addon/socket, default port 7777.
GDScript LSP toolsGodot LSP enabled on port 6005.
DAP debugging toolsGodot DAP enabled on port 6006.
Asset store/provider toolsNetwork access and provider availability.

Add the Godot Plugins

Install from your Godot project folder:

bash
curl -sL https://raw.githubusercontent.com/HaD0Yun/Gopeak-godot-mcp/main/install-addon.sh | bash

PowerShell:

powershell
iwr https://raw.githubusercontent.com/HaD0Yun/Gopeak-godot-mcp/main/install-addon.ps1 -UseBasicParsing | iex

Then enable plugins in Project Settings → Plugins:

  • godot_mcp_editor for bridge-backed scene/resource tools
  • godot_mcp_runtime for runtime inspection, screenshots, and input workflows

Optional shell hooks for update notifications are opt-in:

bash
gopeak setup

gopeak setup only modifies supported bash/zsh rc files when you run it explicitly. npm install does not install shell hooks automatically.


Tool Profiles

GoPeak supports three exposure profiles:

ProfileUse when
compactDefault. Trusted core tools plus dynamic groups activated on demand.
fullCompatibility/audit mode for the full legacy surface.
legacyOlder config alias with the same exposed behavior as full.

Set either GOPEAK_TOOL_PROFILE or the fallback alias MCP_TOOL_PROFILE.

Dynamic groups

In compact mode, search with tool.catalog; matching groups auto-activate. You can also manage groups directly with tool.groups.

Common groups:

GroupStatusNotes
runtimeoptional-runtimeLive scene tree, properties, method calls, metrics. Requires runtime addon/socket.
testingoptional-runtimeScreenshots, viewport capture, input injection. Requires runtime/editor setup.
lspoptional-lspDiagnostics, completion, hover, symbols. Requires Godot LSP on port 6005.
dapoptional-dapBreakpoints, stepping, stack traces. Requires Godot DAP on port 6006.
asset_storeoptional-networkExternal CC0 asset search/download. Network/provider dependent.
class_advancedtrusted-staticClassDB/inheritance discovery backed by static engine metadata.
tilemapaudit-requiredMust account for Godot 4.3+ TileMapLayer behavior before promotion.
mutation groupsaudit-requiredScene/resource/script/settings/signal/autoload/import/audio/navigation/theme/animation groups need fixture evidence before being marketed as fully trusted.
intent_trackingworkflow-layerWorkflow memory/handoff helpers, not a Godot engine primitive. Keep opt-in.

If your MCP client does not refresh after activation, reconnect the client or call the newly activated tool once to force a fresh tools/list round-trip.

GoPeak also uses cursor-based pagination for tools/list so large profiles are not dumped into context at once. Tune it with GOPEAK_TOOLS_PAGE_SIZE when needed.


Typed Godot Values

Bridge-backed scene tools such as add_node and set_node_properties accept common vector payloads for typed properties:

json
{
  "position": { "type": "Vector2", "x": 100, "y": 200 },
  "scale": { "type": "Vector2", "x": 2, "y": 2 }
}

Plain { "x": 100, "y": 200 } and [100, 200] are also coerced for common Vector2 fields, but tagged values are safest across tools.


Useful Commands

bash
# run from npm
npx -y gopeak

# install globally
npm install -g gopeak

# run from source
git clone https://github.com/HaD0Yun/Gopeak-godot-mcp.git
cd Gopeak-godot-mcp
npm install
npm run build
node build/index.js

# local verification
npm run ci
npm run test:dynamic-groups
npm run test:metadata
npm run test:packaging

CLI bin names:

  • gopeak
  • godot-mcp

Environment & Ports

NamePurposeDefault
GOPEAK_TOOL_PROFILETool exposure profile: compact, full, legacycompact
MCP_TOOL_PROFILEFallback profile env aliascompact
GODOT_PATHExplicit Godot executable pathauto-detect
GODOT_BRIDGE_PORTBridge/Visualizer HTTP+WS port override6505
GOPEAK_BRIDGE_HOSTBridge/Visualizer bind host127.0.0.1
GOPEAK_TOOLS_PAGE_SIZENumber of tools per tools/list page33
GOPEAK_RUNTIME_TIMEOUT_MSRuntime addon command timeout in milliseconds10000
DEBUGEnable server debug logsfalse
LOG_MODERecording mode: lite or fulllite
PortService
6505Unified Godot Bridge + Visualizer server on loopback by default.
6005Godot LSP.
6006Godot DAP.
7777Runtime addon command socket.

Runtime screenshot tools (capture_screenshot, capture_viewport) use a GoPeak-managed temporary PNG file when the runtime addon supports output_path, then return normal MCP image content. Older runtime addons that do not receive an output_path continue to return inline base64 screenshots.


Troubleshooting

  • Godot not found → set GODOT_PATH.
  • No MCP tools visible → restart your MCP client.
  • Need a hidden tool → search with tool.catalog or activate a group with tool.groups.
  • Project path invalid → confirm project.godot exists.
  • Runtime tools not working → install/enable the runtime addon and check port 7777.
  • Runtime screenshots time out → update the runtime addon so screenshot commands support the managed output_path flow. For slow runtime responses, raise GOPEAK_RUNTIME_TIMEOUT_MS; older addons may still time out on large inline base64 screenshots.
  • Editor bridge disconnected → stop duplicate gopeak/MCP servers that may already own bridge port 6505; get_editor_status reports bridge startup errors such as EADDRINUSE.

Migration & Deprecation Policy

GoPeak treats compact as the safe default and full/legacy as compatibility profiles. Future hide, remove, rename, or API-contract changes must include:

  1. old → new mapping or an explicit no-replacement note;
  2. profile impact (compact, full, legacy, or opt-in group);
  3. alias window and planned removal timing;
  4. README/docs and release-note updates;
  5. verification proving tools/list exposure and alias behavior;
  6. migration prompt examples for common Godot workflows.

Current stance: legacy tool names and compact aliases remain supported. Optional external groups (runtime, testing, lsp, dap, asset_store) are setup-gated, not always-available core behavior.

Full policy: docs/migration-policy.md.


More Docs


License & Credits

MIT — see LICENSE.

Installation

TypingMind
Prerequisites:

Node.js 18+

{
  "mcpServers": {
    "had0yun-gopeak-godot-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "gopeak"
      ]
    }
  }
}

Use GoPeak MCP with multiple AI models

TypingMind connects MCP tools at the workspace level, so once GoPeak is connected, you can use it with different AI models in TypingMind instead of setting it up separately for each model. This MCP runs locally through the TypingMind MCP connector on your device.

Setup guide to use the local connector

Use this when the MCP server needs access to local files, apps, or private resources on your computer.

1

Open the MCP settings

In TypingMind, go to Settings, Advanced Settings, then Model Context Protocol and choose Setup Connector.

  1. Open TypingMind in your browser.
  2. Click the Settings icon.
  3. Go to Advanced Settings.
  4. Open the Model Context Protocol section.
  5. Click Setup Connector and choose This Device.
TypingMind MCP connector setup screen with This Device selected
2

Run the connector command

Choose This Device, copy the command from TypingMind, and run it in Terminal. Keep the process running while you use MCP.

  1. Copy the setup command shown by TypingMind.
  2. Open Terminal on macOS or Windows Terminal on Windows.
  3. Paste and run the command.
  4. Approve the package install if Terminal asks you to proceed.
  5. Keep the Terminal window running while using MCP tools.
3

Add GoPeak as a server

When the connector status is Ready, click Edit Servers and paste the MCP server configuration.

  1. Wait until the connector status shows Ready.
  2. Click Edit Servers.
  3. Paste the GoPeak MCP server configuration.
  4. Save the server list.
  5. Refresh if you want to confirm the connector is still ready.
TypingMind MCP settings showing active server and Edit Servers button
{
  "mcpServers": {
    "had0yun-gopeak-godot-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "gopeak"
      ]
    }
  }
}
4

Use it across models

Save the server list, open Plugins, enable the GoPeak MCP tools, then select any supported AI model in TypingMind and use the tools in chat or assign them to an AI agent.

  1. Open the Plugins page in TypingMind.
  2. Enable the GoPeak MCP tools.
  3. Start a chat and choose the AI model you want to use.
  4. Use the MCP tools in chat or assign them to an AI agent.
  5. Switch to another AI model whenever needed without reconnecting MCP.
TypingMind chat using enabled MCP tools with a selected AI model
Can you use GoPeak to help me with this task?
GoPeak
Sure. I read it.
Here is what I found using GoPeak.

Frequently asked questions

What is the GoPeak MCP server used for?

GoPeak is an MCP server that lets compatible AI clients connect to external tools and context. In TypingMind, you can add this MCP server once and make its tools available in your AI workspace.

Can I use GoPeak MCP with multiple AI models in TypingMind?

Yes. TypingMind connects MCP tools at the workspace level, so you can use GoPeak with different AI models such as Claude, ChatGPT, Gemini, or other models you have configured in TypingMind without setting up the MCP server separately for each model.

Why use GoPeak MCP with TypingMind?

TypingMind is one of the best frontends for LLM chat because it brings multiple AI models, prompts, plugins, AI agents, API keys, and MCP tools into one workspace. With GoPeak connected, you can use its MCP tools across your preferred models while keeping your chat workflow organized in TypingMind.

How do I connect GoPeak MCP to TypingMind?

GoPeak runs through the TypingMind local MCP connector. This is best when the MCP server needs access to local files, desktop apps, command-line tools, or private resources on your computer.

What tools does GoPeak MCP provide in TypingMind?

GoPeak exposes 33 MCP tools that can be enabled from the TypingMind Plugins page and used in chat or assigned to AI agents.

Do I need to share my API keys with TypingMind to use GoPeak MCP?

No. TypingMind is local-first and lets you keep your model providers, API keys, prompts, and MCP configuration under your control. If GoPeak requires authentication, add the required headers, OAuth settings, or local configuration for that MCP server when you create the connection.

Related MCP Servers

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇