uc-mcp-proxy

MCP stdio-to-Streamable-HTTP proxy with Databricks OAuth.

Lets any MCP client that speaks stdio (e.g. Claude Desktop, Claude Code) connect to a remote Streamable HTTP MCP server hosted on Databricks Apps — handling OAuth authentication automatically.

Installation

# Run directly (no install needed)
uvx uc-mcp-proxy --url https://<workspace>.databricks.com/apps/<app>/mcp

# Or install globally
uv tool install uc-mcp-proxy

Requires Python 3.10+.

Usage

Claude Desktop / Claude Code (.mcp.json)

Add to your MCP client configuration:

{
  "mcpServers": {
    "unity-catalog": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "uc-mcp-proxy",
        "--url", "https://<workspace>.databricks.com/apps/<app>/mcp"
      ]
    }
  }
}

CLI

uc-mcp-proxy --url <MCP_SERVER_URL> [--profile <DATABRICKS_PROFILE>] [--auth-type <AUTH_TYPE>]

| Flag | Description | |------|-------------| | --url | (required) Remote MCP server URL | | --profile | Databricks CLI profile name (uses default if omitted) | | --auth-type | Databricks auth type, e.g. databricks-cli |

How It Works

1. Starts an MCP stdio server (stdin/stdout)
2. Connects to the remote MCP server via Streamable HTTP
3. Injects a fresh Databricks OAuth token on every HTTP request
4. Bridges messages bidirectionally between the two transports

OAuth Authentication

Authentication is handled by the Databricks SDK, which supports multiple auth methods:

• Databricks CLI (databricks-cli) — uses tokens from ~/.databrickscfg
• OAuth U2M — browser-based login flow
• PAT — personal access tokens
• Azure / GCP / AWS — cloud-native identity

The SDK auto-detects the method, or you can force one with --auth-type.

Development

uv sync                        # install dependencies
uv run pytest -m unit -v       # run unit tests
uv run pytest -m integration -v # run integration tests
uv build                       # build package

License

MIT