Connect external IDEs to managed MCP servers

You can connect external AI assistants and IDEs that support MCP to Databricks managed MCP servers. This lets you use Databricks data and tools directly in your development environment.

Prerequisites

Before connecting external tools:

• Generate a Personal Access Token (PAT). See Personal access tokens.

• Identify your MCP server URLs:

  • For UC functions: https://<workspace-hostname>/api/2.0/mcp/functions/{catalog}/{schema}
  • For vector search: https://<workspace-hostname>/api/2.0/mcp/vector-search/{catalog}/{schema}
  • For Genie spaces: https://<workspace-hostname>/api/2.0/mcp/genie/{genie_space_id}

• Ensure your account has access to the Unity Catalog resources (catalogs, schemas, functions, or vector indexes) you want to use.

Configure Cursor

Cursor supports MCP through its settings configuration. Add the following to your Cursor settings:

The example shows UC Functions, but you can adapt it for any managed MCP server URL path:

{
  "mcpServers": {
    "uc-function-mcp": {
      "type": "streamable-http",
      "url": "https://<your-workspace-hostname>/api/2.0/mcp/functions/{catalog_name}/{schema_name}",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      },
      "note": "Databricks UC function"
    }
  }
}

Claude Desktop

To use Databricks-managed MCP servers in Claude Desktop, use mcp-remote. Add the following configuration to your claude_desktop_config.json file.

The configuration file can be found in the following locations by default:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

The example shows UC Functions, but you can adapt it for any managed MCP server URL path:

{
  "mcpServers": {
    "uc-function-mcp": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://<your-workspace-hostname>/api/2.0/mcp/functions/{catalog_name}/{schema_name}",
        "--header",
        "Authorization: Bearer <YOUR_TOKEN>"
      ]
    }
  }
}

After updating the configuration, restart Claude Desktop for the changes to take effect.

Why can't I connect to custom MCP servers?

Managed MCP servers support Personal Access Token (PAT) authentication, which external tools can easily use. When you configure Cursor or Claude Desktop with your PAT token, they can authenticate directly to Databricks managed MCP servers.

Custom MCP servers hosted on Databricks Apps use OAuth authentication, which requires interactive login flows that external IDEs do not support. This means external tools like Cursor and Claude Desktop cannot connect to custom MCP servers.