gb-studio-agent
Claude or other MCP Agentic integration for GB Studio Automated game asset creation and project management REST API endpoints for scenes, actors, assets, and validation End-to-end workflow with your MCP server
claude mcp add --transport stdio eoinjordan-gb-studio-agent node /absolute/path/to/build/mcp.js \ --env CLAUDE_API_KEY="your-claude-api-key-here" \ --env GBSTUDIO_API_URL="http://localhost:3000"
How to use
This MCP server is a TypeScript/Node.js-based MCP endpoint designed to integrate GB Studio projects with Claude-style LLM automation. It exposes an MCP interface that connects a GB Studio project workflow with generation, testing, and augmentation capabilities powered by Claude/CLOB-like tooling. You can use it to scaffold starter GB Studio templates, run lightweight tests on project logic, and generate or extend game assets and scenes through prompt-driven prompts. Connect a compatible MCP client (such as Claude Desktop) to the server, then issue prompts to scaffold, validate, or iterate on GB Studio projects. The server proxies to a local GB Studio API and can be configured to point to different API URLs if you run GB Studio locally or in a remote environment.
Key capabilities you can leverage through MCP include: initiating scaffold templates for Pong, Pac-Man, Mario Bros-style platformers, and space shooter concepts; triggering automated tests or smoke tests for project logic; and iterating on game design by prompting the tool to generate scenes, actors, assets, and event flows that fit GB Studio’s visual game builder paradigm.
How to install
Prerequisites:
- Node.js v14+ (recommended current LTS) and npm installed.
- Access to a shell/terminal.
Installation steps:
-
Install the MCP server globally (as per README example): npm install -g gbstudio-claude-mcp
-
Verify installation and version: gbstudio-claude-mcp --version
-
Ensure GB Studio REST API is accessible if you’re using the REST-based workflow. By default, the MCP server expects GB Studio API at http://localhost:3000 unless you override with GBSTUDIO_API_URL.
-
Start the MCP server and connect a client: gbstudio-claude-mcp
OR if using MCP stdio mode, implement the node/mcp.js approach described in the MCP docs and point your client to the stdio channel.
-
(Optional) Create an MCP client configuration in your tool to target the server, for example using the mcpServers configuration with command and args as shown in the project’s README.
-
Optionally set environment variables for local development (see .env guidance in the README): CLAUDE_API_KEY=your-claude-api-key-here GBSTUDIO_API_URL=http://localhost:3000
Additional notes
Tips and notes:
- The server is designed to work with MCP clients like Claude Desktop. Ensure your client is configured to the correct mcpServer name (e.g., gbstudio-claude-mcp).
- If you run GB Studio behind a different URL/port, update GBSTUDIO_API_URL accordingly so the MCP server can proxy requests to the local REST API.
- For Clawdbot/Moltbot compatibility, this MCP server is stdio-driven; if your bot expects HTTP+SSE MCP, you may need a bridge or adapter.
- Environment variables are optional but helpful for local development (e.g., toggling verbose logs, API keys, or custom GB Studio endpoints).
- The server’s primary use is to scaffold, validate, and extend GB Studio projects via natural prompts. For best results, craft prompts that clearly specify scenes, actors, assets, and gameplay flow in GB Studio terms.
Related MCP Servers
furi
CLI & API for MCP management
mcp -arangodb
This is a TypeScript-based MCP server that provides database interaction capabilities through ArangoDB. It implements core database operations and allows seamless integration with ArangoDB through MCP tools. You can use it wih Claude app and also extension for VSCode that works with mcp like Cline!
mcp-bundler
Is the MCP configuration too complicated? You can easily share your own simplified setup!
CodeRAG
Advanced graph-based code analysis for AI-assisted software development
wormhole
Wormhole: Collaborative AI Workflow Manager🌀
mcp-agentic-sdlc
A comprehensive framework for managing software development lifecycle with AI agents, combining structured development processes with intelligent workflow management.