MCP

The Deliverect API documentation is also available through our Model Context Protocol (MCP) server, an open standard that lets AI tools securely connect to external sources. Once connected, editors like Cursor and Windsurf, or assistants like Claude Desktop, can search our docs and generate integration code directly in your workflow. If your APIs require authentication, pass headers as query parameters or configure them in your MCP client.

Claude Desktop
Add the relevant configuration to claude_desktop_config.json:

Deliverect Restaurants
{
  "mcpServers": {
    "deliverect-restaurants": {
      "url": "https://developers.deliverect.com/mcp?branch=1.1-restaurants"
    }
  }
}

Cursor
Add the relevant configuration to ~/.cursor/mcp.json:

Deliverect Restaurants
{
  "mcpServers": {
    "deliverect-restaurants": {
      "url": "https://developers.deliverect.com/mcp?branch=1.1-restaurants"
    }
  }
}

Windsurf
Add the relevant configuration to ~/.codeium/windsurf/mcp_config.json:

Deliverect Restaurants
{
  "mcpServers": {
    "deliverect-restaurants": {
      "url": "https://developers.deliverect.com/mcp?branch=1.1-restaurants"
    }
  }
}

Testing Your MCP Setup

Once configured, your AI assistant will have access to the Deliverect API documentation through the MCP server. You can test this with the following steps:

  1. Open your AI editor (Cursor, Windsurf, etc.)
  2. Start a new chat with the AI assistant
  3. Ask about Deliverect - try questions like:
    • "How do I submit an order status update?"
    • "Show me an example of an order payload"
    • "Create a full Product sync payload based on my own POS product model"