mastergo-magic
MasterGo Magic MCP is a standalone MCP (Model Context Protocol) service designed to connect MasterGo design tools with AI models.
claude mcp add --transport stdio mastergo-design-mastergo-magic-mcp npx -y @mastergo/magic-mcp --token=<MG_MCP_TOKEN> --url=https://mastergo.com \ --env API_BASE_URL="https://mastergo.com" \ --env MG_MCP_TOKEN="MasterGo API token"
How to use
MasterGo Magic MCP is a standalone MCP service that enables AI models to retrieve DSL data and other MasterGo design information directly from MasterGo design files. It runs via npx, requiring a MasterGo API token and an API base URL. Once connected, you can leverage the included MCP tools to fetch DSL data, get component documentation links, retrieve metadata, and generate component development workflows for Vue and React. The service is designed to be used from within your MCP-enabled editor or integration, where you can invoke the tool running in the background and pass your token and target API URL. The configuration supports both command line arguments and environment variables for token and API URL, making it straightforward to integrate into your development workflow.
To use the MCP server, provide your MasterGo API token and point the tool at the MasterGo API. The package exposes tools for retrieving DSL data from design files, obtaining component documentation links, pulling metadata, and generating structured component workflows. You can also install the MCP via a marketplace or run it locally using the provided command line options. If you encounter permission errors, ensure you are on a Team Edition or higher MasterGo account and that your design files reside in Team Projects rather than a draft area. The local development flow offers building and running a local MCP instance for testing and extending the service.
How to install
Prerequisites:
- Node.js (recommended latest LTS) installed on your machine
- Internet access to fetch npm packages
- A MasterGo API token with appropriate permissions
Install and run (recommended via npx):
- Ensure Node.js and npm are installed:
- node -v
- npm -v
- Run the MCP server using npx (no local installation required):
npx -y @mastergo/magic-mcp --token=<MG_MCP_TOKEN> --url=https://mastergo.com
If you need to customize the URL, replace with your API URL
Using environment variables instead of command line arguments:
- Set environment variables and run the package via npx: MG_MCP_TOKEN=<YOUR_TOKEN> API_BASE_URL=https://mastergo.com npx -y @mastergo/magic-mcp
Optional local development workflow (build from source):
- Clone the repository and install dependencies yarn
- Build the project yarn build
- Find the built entry point (dist/index.js) and run with your token node dist/index.js --token=<MG_MCP_TOKEN> --url=https://mastergo.com
Notes:
- If you plan to install via Smithery Marketplace, follow the marketplace instructions in the README to connect to your MCP client.
Additional notes
Tips and common issues:
- Ensure your MasterGo account is Team Edition or higher; personal/free accounts may not support MCP access.
- Design files must be located in Team Projects; files in the draft box may not be accessible via MCP.
- If you see a 'no permission' error, verify token validity and that the account has access to the relevant projects.
- When using command line options, you can pass multiple --rule entries to apply several design rules.
- Environment variable approach is useful for automation and editor integrations: MG_MCP_TOKEN for the token and API_BASE_URL for the API base URL.
- The npm package name to install/run is @mastergo/magic-mcp, and the example uses npx to run directly without local installation.
- For local development, you can build from source and run with node against dist/index.js, which allows you to modify and extend MCP tools.
Related MCP Servers
iterm
A Model Context Protocol server that executes commands in the current iTerm session - useful for REPL and CLI assistance
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
editor
MCP Server for Phaser Editor
DoorDash
MCP server from JordanDalton/DoorDash-MCP-Server
mcp
MCP сервер для автоматического создания и развертывания приложений в Timeweb Cloud