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 Agents, and custom functions.

  • 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 Custom Agents.

If you are setting up a third-party coding agent such as Claude Code or Cursor, pair these servers with Databricks AI tools. Managed MCP servers give the agent governed tools to call. AI tools teaches it the Databricks patterns to apply when it writes code.

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.

Server

Use case

URL pattern

OAuth scope

Genie One

Natural-language analytics across your workspace

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

genie

Genie Agent

Natural-language analytics scoped to one Genie Agent

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

genie

AI Search

Retrieval over unstructured documents

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

ai-search

Databricks SQL

Developer queries and data engineering

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

sql

Unity Catalog functions

Predefined SQL logic as tools

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

unity-catalog

Server

Use case

URL pattern

OAuth scope

Genie One

Natural-language analytics across your workspace

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

genie

Genie Agent

Natural-language analytics scoped to one Genie Agent

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

genie

AI Search

Retrieval over unstructured documents

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

ai-search

Databricks SQL

Developer queries and data engineering

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

sql

Unity Catalog functions

Predefined SQL logic as tools

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

unity-catalog

To connect clients such as Cursor, Claude Desktop, or ChatGPT to any of these servers, see Connect MCPs to AI assistants and coding agents.

Genie One MCP vs. Databricks SQL MCP servers​

For analytics use cases, start with the Genie One MCP server. Genie resolves business terms through Genie Ontology, your governed semantic layer, which produces more accurate answers than an agent writing SQL directly against raw tables. Use the Databricks SQL MCP server when you need to run a specific query you already wrote, such as validating syntax or authoring a pipeline.

Tool call arguments vs. _meta parameters​

Databricks managed MCP servers handle parameters in two ways:

  • Tool call arguments: Parameters that an LLM typically generates dynamically based on user input
  • _meta parameters: Configuration parameters that you can preset in your agent code to set behavior deterministically

For the specific _meta parameters each server supports, see that server's page (for example, AI Search or Databricks SQL).

Pricing​

Managed MCP server pricing depends on the type of feature:

Additional resources​

On this page