MCP Integration
The Model Context Protocol (MCP) is the bridge between your AI-powered IDE (like Cursor, Windsurf, or VS Code) and your DYPAI backend. By connecting via MCP, your AI assistant can directly manage your infrastructure, data, and logic using natural language.
Why use MCP?
Instead of switching between your code editor and a web dashboard, MCP allows you to build your backend while you develop your frontend. You can ask your AI to:
- Create Tables: "Add a 'products' table with price and stock columns."
- Build APIs: "Create a new POST endpoint to handle user registrations."
- Setup Workflows: "Integrate WhatsApp notifications when a new order is created."
- Manage Data: "Query the last 10 users who signed up today."
How to Connect
How to connect your IDE via MCP
Connecting your project is a simple three-step process:
Get your Connection Details
Click the MCP button in the top header of your DYPAI project dashboard. This will open a dialog with your unique connection URL and configuration snippets.
Configure your IDE
Copy the JSON configuration snippet and paste it into your IDE's MCP settings.
- Cursor: Go to Settings > Models > MCP and add a new server.
- Claude Desktop: Add the configuration to your
claude_desktop_config.json. - Windsurf: Add it to your AI configuration settings.
Verify Connection
Once added, your AI assistant will have access to a set of tools. You can verify the connection by asking: "What tables are available in my project?"
Available Tools
Once connected, your AI assistant can use a wide range of tools to build and manage your project. For a detailed list of what you can ask the AI to do, see our Available Tools page.
Security & Privacy
MCP connections are secured using project-specific tokens.
- Scoping: The AI only sees the project you've connected.
- Auditing: Every action performed by the AI via MCP is logged in your project's history.
- Control: You can regenerate or revoke MCP tokens at any time from the dashboard.