teamcenter-mcp-server-test

Teamcenter MCP Server

Universal MCP server for integrating AI assistants with Teamcenter Knowledge Base APIs with Azure AD authentication support.

📦 Live on PyPI: https://pypi.org/project/teamcenter-mcp-server-test/

✨ What's New in v0.1.2

• 🔐 Azure AD Authentication - Connect to real Teamcenter APIs
• 🔄 Hybrid Mode - Seamless switching between localhost mock and production
• 🌍 Environment Variables - Configure via TEAMCENTER_API_HOST
• 🛡️ Secure - Uses cached Azure AD cookies, no secrets in code

Quick Start (Just Copy & Paste)

🚀 Production Mode (Azure AD)

Add to ~/.continue/config.json:

{
  "experimental": {
    "modelContextProtocolServers": [{
      "transport": {
        "type": "stdio",
        "command": "uvx",
        "args": ["teamcenter-mcp-server-test@0.1.2"],
        "env": {
          "TEAMCENTER_API_HOST": "https://codesentinel.azurewebsites.net"
        }
      }
    }]
  }
}

🔧 Development Mode (Localhost Mock)

Add to ~/.continue/config.json:

{
  "experimental": {
    "modelContextProtocolServers": [{
      "transport": {
        "type": "stdio",
        "command": "uvx",
        "args": ["teamcenter-mcp-server-test@0.1.2"]
      }
    }]
  }
}

VS Code (Production)

Add to .vscode/mcp.json:

{
  "servers": {
    "teamcenter": {
      "type": "stdio",
      "command": "uvx",
      "args": ["teamcenter-mcp-server-test@0.1.2"],
      "env": {
        "TEAMCENTER_API_HOST": "https://codesentinel.azurewebsites.net"
      }
    }
  }
}

JetBrains IDEs (Production)

Add to ~/.mcp.json:

{
  "mcpServers": {
    "teamcenter": {
      "command": "uvx",
      "args": ["teamcenter-mcp-server-test@0.1.2"],
      "env": {
        "TEAMCENTER_API_HOST": "https://codesentinel.azurewebsites.net"
      }
    }
  }
}

🔐 Azure AD Authentication Setup

Prerequisites

• Authenticate first using the working Python client:

  # Run this once to cache Azure AD credentials
  python /path/to/easy_auth_client.py ask "test"
  

• Verify authentication works:

  # Check for cached cookie
  ls ~/.teamcenter_easy_auth_cache.json
  

Environment Variables

• TEAMCENTER_API_HOST: API endpoint URL
  • Production: https://codesentinel.azurewebsites.net
  • Development: http://localhost:8000 (default)

📦 Version History

• v0.1.2 (Latest) - Azure AD authentication + hybrid mode
• v0.1.1 - Localhost mock only (legacy)

Usage

→ See USAGE.md for copy & paste examples ←

Quick examples:

• VS Code: @workspace get Teamcenter API documentation for part creation
• Continue.dev: @MCP search for PLM workflow integration documentation

Production Setup

Replace http://localhost:8000 with your real Teamcenter API:

"args": ["teamcenter-mcp-server-test", "--base-url", "https://teamcenter.yourcompany.com"]

Testing

Quick Test

uvx teamcenter-mcp-server-test --version

Demo/Development Setup

Start mock API server:

git clone https://github.com/your-repo/mock-api
cd mock-api
uv run uvicorn main:app --reload

Server runs on http://localhost:8000 - use this URL in configs above.

Development (Advanced)

Click for development setup

Installation

curl -LsSf https://astral.sh/uv/install.sh | sh

Build Package

uv build

Run Tests

uv run pytest tests/ -v

Publishing to PyPI

→ See DEVELOPER.md for release instructions ←

Files Overview

• auth_mcp_stdio.py: Main MCP server
• main.py: Mock API server for development
• pyproject.toml: Package configuration