Integrate with Cursor

This topic explains how to integrate Devart MCP Server with Cursor so that you can interact with your data source in the Cursor app. You can configure the integration by using the MCP Server connection manager or by editing the Cursor configuration file.

Note

Before you start, make sure that MCP Server has an active connection to the data source. For instructions, see Connection configuration.

Use the MCP Server connection manager

1. In Devart MCP Server, open the connection for editing in one of these ways:

  • Double-click the connection.
  • Click the ellipsis (…) and select Edit.

Devart MCP Server window showing a selected connection, with the Edit option highlighted in the three-dot menu.

2. On the Integration tab, under Cursor, click Register.

3. Optional: Modify the connector name that will appear in the Cursor app.

4. Click Save.

Edit MCP Server screen showing the Integration tab, with Cursor registered.

5. Restart Cursor.

Devart MCP Server is now integrated with Cursor, and the connector appears in the Cursor app in Settings > Tools & MCPs > User MCP Servers.

Note

If you have multiple connections configured in Devart MCP Server, repeat the integration process for every connection. Connector names must be unique.

The devart connector in the Cursor app.

Use the Cursor configuration file

1. Open mcp.json, the Cursor configuration file.

Tip

If you can’t locate the configuration file, it may not exist yet. To have it created, in the Cursor app, navigate to Settings > Tools & MCPs > User MCP Servers, then click Add Custom MCP. The mcp.json file opens.

2. Add one of the following objects, depending on the transport protocol used by MCP Server:

  • STDIO

    {
      "mcpServers": {
        "devart": {
          "command": "C:\\Program Files\\Devart\\MCP Server for <data_source_name>\\Devart.AI.McpServer.Odbc.<data_source_name>.exe",
          "args": [
            "run",
            "devart"
          ]
        }
      }
    }
    

    where:

    • devart is the connector name that will appear in Cursor.

    • <data_source_name> is the name of the target data source.

  • HTTP

    {
      "mcpServers": {
        "devart": {
          "command": "npx",
          "args": [
            "-y",
            "mcp-remote",
            "http://localhost:5500/sse"
          ]
        }
      }
    }
    

    where:

    • devart is the connector name that will appear in Cursor.
    • 5500 is the port specified in the MCP settings for the connection.

Notepad++ window showing `mcp.json` with MCP server and preferences settings in JSON format.

3. Save the file.

4. Restart Cursor.

Devart MCP Server is now integrated with Cursor, and devart appears in the Cursor app in Settings > Tools & MCPs > User MCP Servers.

The Cursor app screen showing the highlighted devart connector tool and its permissions.

Add multiple connections

To use multiple connections in Cursor, add another connector entry to the mcpServers object in mcp.json.

{
  "mcpServers": {
    "devart_mysql": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "http://localhost:5500/sse"
      ]
    },
    "devart_pg": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "http://localhost:3306/sse"
      ]
    }
  }
}