Getting Started with Databricks

Connect your OpenMetadata instance to Databricks by registering it as an MCP Service in Unity Catalog. Once registered, Databricks agents and AI assistants (including Databricks Assistant and Mosaic AI Agent Framework agents) can discover and call OpenMetadata’s MCP tools as a governed, auditable Unity Catalog resource. Unlike client-side integrations such as Claude or Cursor, Databricks doesn’t connect directly to the OpenMetadata MCP endpoint from a local config file. Instead, an admin registers the server once as a Unity Catalog securable, and Databricks proxies and governs every call through its Unity AI Gateway.

Prerequisites

Before you begin, ensure you have:

OpenMetadata v1.8.0+ running.
MCP Server application installed in OpenMetadata.
1. Navigate to Settings > Application and click Add Apps.
2. Install MCP Server application if not already installed.
An OpenMetadata Personal Access Token, or an OpenMetadata user account to sign in with if you want per-user OAuth authentication instead.
A Databricks workspace with Unity Catalog enabled, in a region that supports Model Serving.
A Databricks administrator or service author with these Unity Catalog privileges:
- To create the HTTP connection: CREATE CONNECTION on the target schema.
- To create the MCP Service: USE CATALOG, USE SCHEMA, and CREATE SERVICE on the target schema, plus USE CONNECTION on the HTTP connection.
- To call the MCP Service: EXECUTE on the service, USE CATALOG and USE SCHEMA on its parent, and workspace access.

Configure the MCP Service

Step 1: Create a Unity Catalog HTTP Connection

To create a Unity Catalog HTTP connection in Databricks, follow these steps:

Navigate to Catalog > Connections.
Click Create connection.
Set Connection type to HTTP.
Set Host / MCP server URL to https://<YOUR-OpenMetadata-SERVER>/mcp.
Set Authentication to Bearer token, and paste your OpenMetadata Personal Access Token.
Enable the Is MCP connection checkbox so Databricks treats the endpoint as an MCP server rather than a generic HTTP API.
Click Create to save the connection.

For per-user OAuth authentication, select Dynamic Client Registration. Set the host to https://<YOUR-OPENMETADATA-SERVER> and the base path to /mcp. Databricks discovers /.well-known/oauth-authorization-server and registers its client at /mcp/register. If your policy requires OAuth U2M instead, register a confidential client and allowlist https://<YOUR-DATABRICKS-WORKSPACE>/login/oauth/http.html:

curl -sS -X POST "https://<YOUR-OPENMETADATA-SERVER>/mcp/register" \
  -H "Content-Type: application/json" \
  -d '{
    "client_name": "Databricks MCP connection",
    "redirect_uris": ["https://<YOUR-DATABRICKS-WORKSPACE>/login/oauth/http.html"],
    "grant_types": ["authorization_code", "refresh_token"],
    "response_types": ["code"],
    "token_endpoint_auth_method": "client_secret_post",
    "scope": "openid profile email"
  }'

Configure Databricks with:

The client_id and client_secret returned when you register the client.
The scopes listed in /.well-known/oauth-authorization-server.
https://<YOUR-OPENMETADATA-SERVER>/mcp/authorize as the authorization endpoint and https://<YOUR-OPENMETADATA-SERVER>/mcp/token as the token endpoint.
header_and_body as the credential exchange method.

OpenMetadata supports only the authorization_code and refresh_token grants, so don’t select OAuth M2M or client_credentials. For a shared, non-interactive credential, use a Personal Access Token instead. For more information, see OAuth 2.0 Authentication. If you selected Dynamic Client Registration or OAuth U2M, each user must authenticate individually before they can call the service:

Open the connection’s detail page in Catalog Explorer.
Click Login.
Complete the OAuth consent flow against your OpenMetadata instance.

Bearer token skips this step: it uses a single shared credential set once, at connection creation, so no per-user login is required.

Step 2: Register the MCP Service

Using the Databricks UI

To register the MCP service in Databricks, follow these steps:

In the left sidebar, select AI Gateway > MCPs > Register MCP Server. Alternatively, go to Catalog, open the target schema, and select Create > MCP Service, which opens the same form.
Set Catalog and Schema to where the MCP Service will live as a Unity Catalog securable.
Set MCP Service name, for example openmetadata (this can’t be changed after creation).
Set Connection to the HTTP connection you created in Step 1.
Optional: Set Tools to the OpenMetadata MCP tools you want to expose. Leave unrestricted to expose all of them. See the MCP Tools Reference for the full list.
Click Create to register the service.

Using the REST API

databricks api post \
  "/api/2.1/unity-catalog/mcp-services?parent=schemas/main.default&mcp_service_id=openmetadata" \
  --json '{
    "comment": "OpenMetadata MCP Server",
    "config": {
      "source_connection": {
        "name": "connections/main.default.openmetadata_http"
      },
      "include_tool_selectors": []
    }
  }'

Step 3: Grant EXECUTE Permission

On the MCP Service’s Permissions tab, grant EXECUTE to the users, groups, or service principals that should be able to call OpenMetadata’s tools:

databricks api patch \
  "/api/2.1/unity-catalog/permissions/mcp_service/main.default.openmetadata" \
  --json '{
    "changes": [
      { "principal": "data-team", "add": ["EXECUTE"] }
    ]
  }'

Grant EXECUTE on the MCP Service itself. Never grant USE CONNECTION on the underlying HTTP connection to end users. That would let them bypass tool selection and call the OpenMetadata MCP endpoint directly with the shared credential.

Step 4: Connect an Agent or Client

After getting registered, the MCP Service is reachable at:

https://<YOUR-DATABRICKS-WORKSPACE>/ai-gateway/mcp-services/<catalog>.<schema>.<service-name>

Mosaic AI Agent Framework agents can add it as a tool source directly. To verify the connection manually, authenticate with a Databricks token and list the available tools:

TOKEN=$(databricks auth token | jq -r .access_token)
curl -s -X POST \
  "https://<YOUR-DATABRICKS-WORKSPACE>/ai-gateway/mcp-services/main.default.openmetadata" \
  -H "Authorization: Bearer $TOKEN" \
  -H "Accept: application/json, text/event-stream" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'

You should see OpenMetadata’s MCP tools (search_metadata, get_entity_details, and others) in the response. See the MCP Tools Reference for the complete list.

Try It Out

Once your agent has the OpenMetadata MCP Service attached as a tool source, try prompting it:

“What tables do you have access to in OpenMetadata?”

“Find the tables that track customer transactions, engagement metrics, and churn indicators.”

Troubleshooting

If you run into connection issues:

Verify OpenMetadata is running: curl <YOUR-OpenMetadata-SERVER>/api/health.
Check the MCP endpoint: curl <YOUR-OpenMetadata-SERVER>/mcp (should return 401).
Verify MCP Application is installed: Visit <YOUR-OpenMetadata-SERVER>/marketplace/apps/McpApplication.
Confirm the connection is marked as an MCP connection: Open the HTTP connection in Catalog Explorer and check that Is MCP connection is enabled.
Confirm EXECUTE is granted: Check the MCP Service’s Permissions tab for the calling user, group, or service principal.
Token expired: Generate a new PAT and update the HTTP connection’s credentials.

​Getting Started with Databricks

​Prerequisites

​Configure the MCP Service

​Step 1: Create a Unity Catalog HTTP Connection

​Step 2: Register the MCP Service

​Using the Databricks UI

​Using the REST API

​Step 3: Grant EXECUTE Permission

​Step 4: Connect an Agent or Client

​Try It Out

​Troubleshooting

​Additional Resources

Getting Started with Databricks

Prerequisites

Configure the MCP Service

Step 1: Create a Unity Catalog HTTP Connection

Step 2: Register the MCP Service

Using the Databricks UI

Using the REST API

Step 3: Grant EXECUTE Permission

Step 4: Connect an Agent or Client

Try It Out

Troubleshooting

Additional Resources