factory
MCP server from cpage-pivotal/factory-mcp-server
claude mcp add --transport stdio cpage-pivotal-factory-mcp-server node server.js \ --env PORT="3000" \ --env NODE_ENV="production"
How to use
The Factory MCP Server provides a simulated manufacturing supply chain dashboard alongside a natural language query interface powered by MCP. You can interact with the dashboard to monitor inventory, production status, order flow, and supply chain metrics, while also querying the system in natural language to retrieve insights, statuses, and trends. The server exposes endpoints and a chat-like interface for questions such as current work-in-progress, lead times, bottlenecks, and forecasted demand, returning structured results that can be consumed by downstream tooling.
To use the tools, start the server and access the dashboard in your browser or via API calls. Use the natural language query capability to ask questions like: “What is the current inventory risk for component X?”, “Show the production throughput for last 24 hours,” or “Forecast next week’s material shortage risk.” The system will interpret the query, run the appropriate simulated data workflows, and return actionable results alongside visualizations in the dashboard where available.
How to install
Prerequisites:
- Node.js (≥14.x) and npm installed on your system
- Basic familiarity with running command-line programs
Installation steps:
-
Clone the repository: git clone https://example.com/your-repo.git cd your-repo
-
Install dependencies: npm install
-
Configure environment (optional):
- Create a .env file with PORT and other settings if you need to override defaults PORT=3000 NODE_ENV=production
-
Start the server: npm start
or if a different script is used in package.json, use the documented start command
-
Verify:
- Open http://localhost:3000 in your browser
- Access the MCP query interface and the dashboard to verify end-to-end functionality
Additional notes
Notes and tips:
- The MCP server is designed to be self-contained with a simulated dashboard; for production use, consider plugging in real data sources or adapters.
- Environment variables can customize port, environment, and data sources. If you change PORT, update any reverse proxies or client scripts accordingly.
- If natural language queries fail to resemble expected intents, try rephrasing or using more specific terms (e.g., “inventory value for item A” or “throughput for product line B”).
- Check logs for DEBUG or INFO level messages to diagnose issues related to data generation or query parsing.
- Ensure data simulations are seeded if you rely on repeatable test results. The seed value can usually be configured in the environment or a config file.
