windchill
Test bed for a MCP server for Windchill. Running in Docker
claude mcp add --transport stdio cars-10-windchill-mcp-server node dist/server.js \ --env WINDCHILL_BASE_URL="https://your-windchill-base-url" \ --env WINDCHILL_PASSWORD="your-windchill-password" \ --env WINDCHILL_USERNAME="your-windchill-username"
How to use
Windchill MCP Server exposes 64+ tools across 7 specialized agents for interacting with PTC Windchill PLM systems via a standardized MCP JSON-RPC interface. The web UI provides tool discovery, automated form generation based on JSON Schema, and real-time tool execution. You can access the MCP API directly at http://localhost:3000/api/ and manage or explore tools through the Angular-based UI at http://localhost:4200. The server supports dynamic server switching and is designed for integration with AI assistants like Claude Desktop, enabling natural-language workflows that map to Windchill operations such as part management, document handling, change requests, workflows, and project administration. Use the web interface to browse tools by agent (Part, Document, Change, Workflow, Project, DataAdmin, ServerManager) or search across categories, then execute a tool to see results rendered in the UI or return structured JSON-RPC responses for downstream automation.
How to install
Prerequisites:
- Node.js (LTS version) and npm installed on your system
- Git for cloning repositories
- Optional: Docker if you prefer containerized deployment
Installation steps:
-
Clone the repository: git clone <your-repo-url>
-
Navigate into the project: cd windchill-mcp-server
-
Install dependencies: npm install
-
Build the server (if a build step is required): npm run build
-
Prepare environment variables (example): Create a .env file or export variables in your shell with Windchill credentials and base URL: WINDCHILL_USERNAME=your-username WINDCHILL_PASSWORD=your-password WINDCHILL_BASE_URL=https://your-windchill-base-url
-
Run the server (as an MCP node server): npm start
-
Optional: If you prefer Docker deployment, see docs/DOCKER.md for containerized setup and pre-configured scripts.
Additional notes
Notes and tips:
- The server exposes a large set of tools across agents; use the web UI to discover and filter by agent or category.
- Ensure Windchill credentials have appropriate permissions for the tools you intend to use.
- If you run into authentication issues, verify that WINDCHILL_BASE_URL is reachable from the host and that any required two-factor or SSO configurations are accounted for in your environment.
- For Claude Desktop integration, enable native stdio support or run in stdio mode if you want stdout/stderr to be the primary I/O channels (see v1.2.0 enhancements in the changelog).
- The MCP API uses JSON-RPC 2.0; client applications should structure requests to methods like tools/list and tools/call with appropriate parameters as illustrated in the documentation.
- When using Docker, ensure Docker is installed and the docker-compose setup corresponds to the docker/docker-compose.yml in the repository and that port mappings (3000 for API, 4200 for UI) do not conflict with other services.
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