Skip to main content
Last updated on

Databricks managed MCP servers

Preview

This feature is in Public Preview.

Databricks managed MCP servers are ready-to-use servers that connect your AI agents to data in Unity Catalog, Databricks AI Search indexes, Genie Spaces, and custom functions. They also connect agents to common software-as-a-service (SaaS) applications like Slack and GitHub.

  • No setup: Databricks hosts the servers and manages authentication.
  • Governed: Unity Catalog enforces permissions, so agents and users access only the tools and data you grant them.
  • Centralized: view, monitor, and manage every server from Unity AI Gateway.

To call these servers from agent code, see Use MCP servers in agents.

Databricks also has ready-to-use MCP Services in the system.ai schema for common SaaS apps (Slack, GitHub, Google Drive, and more). See Databricks-provided MCP Services.

Available managed servers​

Databricks has the following MCP servers that work out of the box. When connecting to managed MCP servers using on-behalf-of user authentication, include the corresponding OAuth scope for each server your application needs to access. For setup instructions, see Authentication methods.

Genie One​

Beta

This feature is in Beta.

URL pattern

OAuth scope

https://<workspace-hostname>/api/2.0/mcp/genie

genie

Ask natural-language data questions about your enterprise data. Genie One searches across Genie Spaces and your Unity Catalog data, then returns a grounded answer with a deep link back to the conversation in the Databricks UI. Read-only.

Genie runs asynchronously. Call the genie_ask tool to start a conversation, then call genie_poll_response until the response is complete. To continue an existing conversation, pass the previous conversation_id to genie_ask.

Genie Space​

URL pattern

OAuth scope

https://<workspace-hostname>/api/2.0/mcp/genie/{genie_space_id}

genie

Query a single Genie Space with natural language. Read-only.

The Genie Space server invokes Genie as a tool, so it doesn't pass conversation history to the Genie API. To preserve history, use Genie in a multi-agent system.

URL pattern

OAuth scope

https://<workspace-hostname>/api/2.0/mcp/ai-search/{catalog}/{schema}/{index_name}

ai-search

Query AI Search indexes to find relevant documents. Requires Databricks managed embeddings.

AI Search was formerly Vector Search. The previous /api/2.0/mcp/vector-search/ URL prefix and vector-search scope still work.

Databricks SQL​

URL pattern

OAuth scope

https://<workspace-hostname>/api/2.0/mcp/sql

sql

Run AI-generated SQL to create data pipelines from AI coding tools. Read and write.

Databricks SQL runs asynchronously: call the tool to start, then poll until the response completes.

Unity Catalog functions​

URL pattern

OAuth scope

https://<workspace-hostname>/api/2.0/mcp/functions/{catalog}/{schema}/{function_name}

unity-catalog

Run Unity Catalog functions as predefined SQL tools.

Example: a customer-support agent​

Connect one agent to multiple managed MCP servers:

  • AI Search (/api/2.0/mcp/ai-search/prod/customer_support) — search support tickets and documentation.
  • Genie Space (/api/2.0/mcp/genie/{billing_space_id}) — query billing data and customer information.
  • Unity Catalog functions (/api/2.0/mcp/functions/prod/billing) — run custom functions for account lookups.

This gives the agent unstructured data (tickets), structured data (billing), and custom business logic in one place.

Additional resources​

On this page