Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
Important
This feature is in Public Preview.
Azure Databricks managed MCP servers are ready-to-use servers that connect your AI agents to data in Unity Catalog, Azure Databricks AI Search indexes, Genie Agents, and custom functions.
- No setup: Azure 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 Azure Databricks AI tools. Managed MCP servers give the agent governed tools to call. AI tools teaches it the Azure 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 |
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
Azure Databricks managed MCP servers handle parameters in two ways:
- Tool call arguments: Parameters that an LLM typically generates dynamically based on user input
_metaparameters: 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:
Unity Catalog functions use serverless general compute pricing.
Genie Agents use serverless SQL compute pricing.
Databricks SQL servers use Databricks SQL pricing.
AI Search indexes use AI Search pricing.
Additional resources
- Use MCP servers in Custom Agents to call managed MCP servers from agent code.
- Connect MCPs to AI assistants and coding agents to connect clients like Cursor and Claude Desktop.
- Agent skills for AI coding assistants to teach coding agents Azure Databricks development patterns.