Databricks Unity Catalog MCP Server
Provides read-only access to Databricks Unity Catalog and Jobs, allowing AI agents to query tables, inspect job configurations, and retrieve metadata from a Databricks workspace.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Databricks Unity Catalog MCP Server list all tables in my Unity Catalog"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Databricks Unity Catalog MCP Server
Access your Databricks workspace through Claude and other LLMs. Query Unity Catalog tables, inspect jobs, and retrieve detailed metadata—all through the Model Context Protocol.
Built on the Databricks SDK to provide read-only access to your workspace through the Model Context Protocol. Powered by FastMCP with async/aiohttp for efficient parallel data retrieval.
Read more about our vision and use cases here.
Table of Contents
Related MCP server: Databricks MCP Server
Features
Capabilities
What you can do:
Ask Claude to find tables in your Unity Catalog
Inspect job configurations and recent runs
Generate queries based on your schema
Limitations
What you can't do:
Modify tables or jobs (read-only by design)
Execute queries directly (retrieves metadata only)
Available Tools
Unity Catalog
Tool | Description | Parameters |
| List all tables across catalogs and schemas | None |
| Retrieve table descriptions, columns, and metadata |
|
Jobs
Tool | Description | Parameters |
| List all workspace jobs with IDs and names | None |
| Get job settings, configurations, and tasks |
|
| Fetch recent run history with duration, parameters, and results |
|
Quick Start
Prerequisites:
Docker Desktop installed and running
Databricks workspace access (host URL and access token)
Installation
Choose your editor and follow the configuration steps:
Step 1: Add the following configuration to .cursor/mcp.json:
{
"mcpServers": {
"databricks": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"DATABRICKS_HOST",
"-e",
"DATABRICKS_TOKEN",
"ghcr.io/revodatanl/databricks-mcp-server:latest"
],
"env": {
"DATABRICKS_HOST": "${env:DATABRICKS_HOST}",
"DATABRICKS_TOKEN": "${env:DATABRICKS_TOKEN}"
}
}
}
}Note: You can either use environment variable references (
${env:VARIABLE}) or hardcode the values as strings directly in the configuration.
Step 2: Create a .env file in your project root with your credentials:
DATABRICKS_HOST=your-workspace-url
DATABRICKS_TOKEN=your-access-tokenStep 3: Restart Cursor to load the MCP server.
Step 4: Use the cursor rules to enhance your Databricks development workflow.
Learn more about MCP in Cursor
Step 1: Add the following configuration to .continue/mcpServers/databricks-mcp.yaml:
name: databricks_mcp_server
version: 0.1.3
schema: v1
mcpServers:
- name: databricks_mcp_server
command: docker
args:
- run
- -i
- --rm
- -e
- DATABRICKS_HOST=${{ inputs.DATABRICKS_HOST }}
- -e
- DATABRICKS_TOKEN=${{ inputs.DATABRICKS_TOKEN }}
- ghcr.io/revodatanl/databricks-mcp-server:latestStep 2: Set your credentials either:
On the Continue.dev website (recommended for security)
Or in a
.envfile in your project root:DATABRICKS_HOST=your-workspace-url DATABRICKS_TOKEN=your-access-token
Step 3: Restart your editor to load the MCP server.
Step 4: Use the Continue.dev rules to enhance your Databricks development workflow.
Learn more about MCP in Continue.dev
Local Development
For contributors and developers who want to run the server locally:
Setup
Install uv - Fast Python package installer Follow the installation guide
Clone the repository
git clone https://github.com/revodatanl/databricks-mcp-server.git cd databricks-mcp-serverInstall dependencies
uv syncSet environment variables
export DATABRICKS_HOST=your-workspace-url export DATABRICKS_TOKEN=your-access-tokenRun the server
uv run databricks-mcp
License
MIT License - see LICENSE.md for details.
Available Tools
8 toolsfind-tables-by-nameA
Find tables by name in Unity Catalog using fuzzy matching with in-memory caching.
Args:
- search_term: The search term to find tables by
- limit: The maximum number of results to return
- force_refresh: If True, bypass the cache and fetch fresh data from the API
Inputs: search_term (required), limit, force_refresh. Returns: error, content, success
get-catalogsB
Retrieve a list of all available catalogs in the Databricks workspace
Inputs: no named parameters. Returns: error, content, success
get-job-detailsC
Get job details like settings and tasks by the job id.
Args:
- List of integer job IDs
Inputs: job_ids (required). Returns: error, content, success
get-job-runsC
Get recent job runs by job id.
Args:
- job_ids: List of job ids for which to get the runs
- amount: Amount of runs to get per job, sorted by most recent
Inputs: job_ids (required), n_recent. Returns: error, content, success
get-jobsA
Retrieve a list of all jobs in the workspace
Inputs: no named parameters. Returns: error, content, success
get-schemas-in-catalogsB
Retrieve a list of all available schemas in the given catalogs.
Args:
- catalog_names: List of catalog names to get schemas for
Inputs: catalog_names (required). Returns: error, content, success
get-tables-detailsA
Retrieve detailed information for the given tables.
Args:
- full_table_names: List of full table names in the format catalog.schema.table to get details for
Inputs: full_table_names (required). Returns: error, content, success
get-tables-in-catalogs-schemasC
Retrieve a list of all tables in the given catalogs and schemas.
Args:
- catalog_schemas: List of catalog.schema strings to get tables for
Inputs: catalog_schemas (required). Returns: error, content, success
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 8 tools
The tools generally target distinct resources and actions, but the three table-oriented tools (list by schema, fuzzy name search, and details) could be confused at a glance despite clear descriptions. Job-related tools are clearly separated between listing, details, and runs.
Most names follow a consistent kebab-case get-<resource> pattern, with only find-tables-by-name deviating in verb choice. The convention is still readable and predictable overall.
Eight tools are well-scoped for a read-only Unity Catalog and jobs exploration server. Each tool covers a distinct browsing or detail-retrieval need without excessive surface area.
The surface supports catalog, schema, table, and job discovery plus details and runs, but it is entirely read-only and lacks operations such as creating/updating/deleting assets, triggering jobs, or accessing permissions/lineage. These gaps limit it to exploration workflows rather than full lifecycle management.
Maintenance
Related MCP Connectors
Your Databricks Lakehouse in natural language: run SQL on your SQL warehouses, track long-running qu
Query your warehouse or a CSV with Claude/ChatGPT over MCP, governed by table-level ACL + audit.
Query 40 databases from Claude, ChatGPT, or Cursor — on any device. Read-only, encrypted, audited.
The grounded data layer for any LLM: governed SQL, metrics, lineage and catalog over your data.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables AI assistants like Claude to interact with Databricks workspaces through secure OAuth authentication. Supports custom prompts, tools for cluster management, SQL execution, and job operations via the Databricks SDK.19-
- FlicenseNot gradedqualityNot gradedmaintenanceEnables AI assistants like Claude to interact with Databricks workspaces through secure OAuth authentication. Supports custom prompts, tools for cluster management, SQL execution, and job operations via the Databricks SDK.-
- FlicenseNot gradedqualityDmaintenanceEnables AI assistants like Claude to interact with Databricks workspaces through custom prompts and tools. Supports running SQL queries, managing clusters, creating jobs, and accessing workspace resources via the Databricks SDK.2-
- FlicenseNot gradedqualityDmaintenanceEnables AI assistants like Claude to interact with Databricks workspaces through secure OAuth authentication. Supports custom prompts, tools for cluster management, SQL execution, and job operations via the Databricks SDK.-