TengineAI Documentation

Everything you need to integrate TengineAI with your favorite AI tools and build powerful applications.

What is the Model Context Protocol (MCP)?

The Model Context Protocol (MCP) is an open standard that provides a standardized way for AI models and assistants to interact with external tools and data sources. Think of it as a universal interface that allows AI to access APIs and services through a consistent protocol.

MCP uses a client-server architecture where MCP clients (like Claude Desktop, Cursor, VS Code, or programmatic clients using MCP SDKs) connect to MCP servers (like TengineAI) to access tools and resources.

How MCP Works

  1. An MCP client (desktop app or programmatic) connects to an MCP server
  2. The server provides a list of available tools and resources
  3. The client can request tools to be executed
  4. The server handles authentication, API calls, and returns results
  5. The AI assistant uses the results to complete tasks

Key Terminology

  • Remote MCP Server: An MCP server hosted in the cloud (like TengineAI) that clients connect to via URL
  • MCP Client: Applications like Claude Desktop, Cursor, VS Code, or programmatic clients using MCP SDKs
  • Tools: Functions that the MCP server provides, such as "send email" or "check calendar"
  • OAuth Tools: Tools that require user authorization (e.g., Gmail, Calendar) with secure token storage
  • API Key Tools: Tools that use project-scoped API keys (e.g., OpenWeather) for simpler authentication

How to Connect to TengineAI

You can connect to TengineAI in two ways:

1. Via MCP Clients

Connect through desktop applications like Claude Desktop (using Custom Connectors in paid plans), Cursor, and VS Code. These applications act as MCP clients and can connect to TengineAI's remote MCP server.

2. Programmatically

Use official MCP SDKs (Python, Java, C#) to connect to TengineAI from your code, enabling you to integrate MCP functionality directly into your applications. This allows you to build custom integrations and make your own APIs accessible to AI assistants.

Client Setup Guides

Need Help?

Can't find what you're looking for? Our support team is here to help.