Tulip MCP Server

Tulip MCP Server is an MCP server that provides access to the Tulip manufacturing platform API for large language model clients. Built for industrial engineers, operations managers, and software developers working with frontline operations, it connects MCP-compatible tools to Tulip instances. The server enables AI assistants to inspect and interact with manufacturing resources such as tables, records, machines, plant floor stations, interfaces, and user directories. By bridging LLMs with industrial data, users can query active production metrics, retrieve table schemas, track equipment statuses, and inspect shop-floor configurations directly through conversation. It defaults to safe, read-only and table-focused operations, while allowing administrators to configure granular tool access across read, write, and administrative capabilities. The server also supports multiple workspace instances, letting enterprise teams manage and compare distinct manufacturing environments from a single LLM interface.

Category: CRM, ERP & E-commerce

Tags: erp, industrial, iot, manufacturing, tulip

Visit Tulip MCP Server

How to install and configure Tulip MCP Server

1. Ensure Node.js is installed on your system. 2. Create a .env file containing your Tulip credentials: env TULIP_API_KEY=your_api_key_here TULIP_API_SECRET=your_api_secret_here TULIP_BASE_URL=https://your-instance.tulip.co TULIP_WORKSPACE_ID=your_workspace_id_if_using_account_key 3. Copy the full absolute path to your .env file. 4. Open your Claude Desktop configuration file (claude_desktop_config.json) and add the server to mcpServers: json { "mcpServers": { "tulip-mcp": { "command": "npx", "args": [ "@tulip/mcp-server", "--env", "/full/path/to/your/.env" ] } } } 5. Restart Claude Desktop or configure the command in Cursor.

What you can do with Tulip MCP Server

• Querying Tulip operational tables to analyze real-time shop-floor records, product assemblies, and inventory levels directly from an LLM. - Auditing manufacturing station and station-group configurations to confirm active assignments and machine-operator pairings across plants. - Inspecting machine telemetry and interface statuses to diagnose production bottlenecks or equipment downtime without switching dashboards. - Comparing configurations and records across multiple enterprise workspaces simultaneously by deploying dedicated server instances for production and QA.

Key facts

• https://github.com/tulip/tulip-mcp
• CRM, ERP & E-commerce, Data & Analytics
• erp, industrial, iot, manufacturing, tulip

Part of MCP Servers

How do I install Tulip MCP Server?

You can run the server directly using Node.js and npx. First, create an environment file containing your Tulip API key, API secret, and base URL. Then, configure your MCP client, such as Claude Desktop or Cursor, to execute npx with the package name @tulip/mcp-server and provide the absolute path to your environment file via the --env argument.

What can Tulip MCP Server do?

The server connects AI clients to the Tulip manufacturing API. It enables assistants to read and manage records in Tulip tables, view connected machines, query station statuses, inspect interface configurations, and manage user accounts across frontline manufacturing deployments.

Which MCP clients work with Tulip MCP Server?

The server works with any standard Model Context Protocol client capable of launching local command-line processes. Verified clients include Claude Desktop and Cursor, which allow you to specify the launch arguments and environment file paths directly in their respective configuration files.

How do I limit which tools the server exposes?

You can control tool availability by setting the ENABLED_TOOLS variable in your environment file. You can define specific tools like listTables, security categories like read-only and write, or resource types like table, machine, and station. By default, the server only activates read-only and table tools.

Can I connect multiple Tulip workspaces?

Yes. You can create separate environment files for each workspace and define multiple server entries in your MCP client configuration, such as tulip-production and tulip-qa. This setup allows your AI client to access and query data across several instances within the same conversation.