databricks-utils-mcp 0.1.0

MCP server for Databricks development and operations

Databricks Utils MCP Server

An MCP (Model Context Protocol) server for Databricks development and operations. Compatible with any MCP client: Claude Code, Claude Desktop, Cursor, and others.

Covers ten areas:

• Unity Catalog -- browse catalogs, schemas, tables, volumes, and functions; describe table schemas; sample table data
• SQL -- execute SQL statements; manage SQL warehouses (list, start, stop)
• Query History -- list recent queries; get full query details
• Jobs -- list, inspect, trigger, cancel, and repair job runs
• Clusters -- list (with filtering and paging), inspect, start, and terminate clusters
• Pipelines -- list, inspect, start, and stop Delta Live Tables pipelines
• Workspace -- browse workspace directories; export notebooks
• Files -- list and read files from DBFS and Unity Catalog Volumes
• Secrets -- list secret scopes and key names (values are never returned)
• Permissions -- get Unity Catalog grants and workspace object ACLs

Authentication uses the Databricks SDK's standard credential chain, which auto-discovers credentials from ~/.databrickscfg profiles, environment variables, or Azure CLI. Raw tokens are never passed as tool parameters. See Authentication below.

Requirements

• uv
• A Databricks workspace with a personal access token, OAuth config, or Azure CLI session

Installation

macOS

brew install uv

Linux

curl -LsSf https://astral.sh/uv/install.sh | sh

Windows

winget install --id=astral-sh.uv

Configuration

Claude Code users:

claude mcp add --scope user databricks-utils -- uvx databricks-utils-mcp

For other MCP clients, add the following to your server configuration:

{
  "mcpServers": {
    "databricks-utils": {
      "command": "uvx",
      "args": ["databricks-utils-mcp"]
    }
  }
}

Restart your MCP client after adding the server.

Installing from source

git clone https://github.com/BrianDeacon/databricks-utils-mcp
cd databricks-utils-mcp
uv sync

Then configure with the cloned path:

{
  "mcpServers": {
    "databricks-utils": {
      "command": "uv",
      "args": ["run", "--directory", "/path/to/databricks-utils-mcp", "databricks-utils-mcp"]
    }
  }
}

---

Authentication

All tools use the Databricks SDK's unified authentication. By default, the SDK checks (in order): environment variables, ~/.databrickscfg default profile, Azure CLI, and other standard credential sources.

Every tool accepts three optional parameters to override the default:

Parameter	Description
profile	Name of a ~/.databrickscfg profile (provides both host and credentials)
host	Workspace URL override (credentials still resolved via the standard chain)
token_env_var	Name of an environment variable holding a PAT. The tool reads the value locally; the token itself is never sent as a parameter.

Example ~/.databrickscfg

[DEFAULT]
host  = https://adb-1234567890.0.azuredatabricks.net/
token = dapi...

[staging]
host      = https://adb-0987654321.0.azuredatabricks.net/
auth_type = databricks-cli

With this config, tools use the DEFAULT profile automatically. Pass profile="staging" to target the staging workspace instead.

---

Unity Catalog Tools

catalog_list_catalogs

List all Unity Catalog catalogs accessible to the current user. Returns a sorted JSON array of catalog names.

catalog_list_schemas

Parameter	Type	Required	Description
catalog	string	yes	Catalog name

catalog_list_tables

Parameter	Type	Required	Description
catalog	string	yes	Catalog name
schema	string	yes	Schema name

catalog_describe_table

Returns columns (name, type, comment), table type, storage location, properties, and timestamps.

Parameter	Type	Required	Description
full_name	string	yes	Three-part name: catalog.schema.table

catalog_get_table_sample

Executes SELECT * LIMIT against a table using a SQL warehouse. Returns a JSON array of row objects.

Parameter	Type	Required	Description
full_name	string	yes	Three-part name: catalog.schema.table
limit	integer	no	Number of rows (default 10, max 100)
warehouse_id	string	no	SQL warehouse ID. If omitted, uses the first running warehouse.

catalog_list_volumes

Parameter	Type	Required	Description
catalog	string	yes	Catalog name
schema	string	yes	Schema name

catalog_list_volume_files

Parameter	Type	Required	Description
volume_path	string	yes	Path under /Volumes/catalog/schema/volume/

catalog_list_functions

Parameter	Type	Required	Description
catalog	string	yes	Catalog name
schema	string	yes	Schema name

---

SQL Tools

sql_execute

Execute a SQL statement against a Databricks SQL warehouse. Returns a JSON array of row objects for queries, or a status message for DDL/DML.

Parameter	Type	Required	Description
statement	string	yes	SQL statement to execute
warehouse_id	string	no	SQL warehouse ID. If omitted, uses the first running warehouse.
max_rows	integer	no	Maximum rows to return (default 100, cap 10,000)
catalog	string	no	Default catalog for unqualified names
schema	string	no	Default schema for unqualified names

sql_list_warehouses

List all SQL warehouses. Returns a JSON array with id, name, state, cluster_size, and auto_stop_mins.

sql_get_warehouse

Parameter	Type	Required	Description
warehouse_id	string	yes	SQL warehouse ID

sql_start_warehouse

Start a stopped SQL warehouse. Does not wait for it to finish starting.

Parameter	Type	Required	Description
warehouse_id	string	yes	SQL warehouse ID

sql_stop_warehouse

Stop a running SQL warehouse.

Parameter	Type	Required	Description
warehouse_id	string	yes	SQL warehouse ID

---

Query History Tools

query_history_list

List recent SQL queries. Returns query ID, statement (truncated), status, duration, rows produced, user, and warehouse.

Parameter	Type	Required	Description
warehouse_id	string	no	Filter to a specific warehouse
user_name	string	no	Filter by user
status	string	no	Filter by status: FINISHED, FAILED, CANCELED, RUNNING
max_results	integer	no	Max results (default 25, cap 100)

query_history_get

Get full details for a specific query including the complete statement text, metrics, and error message if failed.

Parameter	Type	Required	Description
query_id	string	yes	Query ID

---

Jobs Tools

jobs_list

Parameter	Type	Required	Description
name	string	no	Filter by job name (substring match)
max_results	integer	no	Max results (default 25)

jobs_get

Returns tasks, schedule, clusters, parameters, tags, and notifications.

Parameter	Type	Required	Description
job_id	integer	yes	Job ID

jobs_list_runs

Parameter	Type	Required	Description
job_id	integer	no	Filter to a specific job
active_only	boolean	no	Only show active (in-progress) runs (default false)
max_results	integer	no	Max results (default 25)

jobs_get_run

Returns per-task states, start/end times, cluster info, error messages, and attempt number.

Parameter	Type	Required	Description
run_id	integer	yes	Run ID

jobs_get_run_output

Returns notebook output, error trace, and logs depending on task type. Only works for single-task runs.

Parameter	Type	Required	Description
run_id	integer	yes	Run ID

jobs_run

Trigger a job run. Does not wait for completion. Returns the run_id for tracking.

Parameter	Type	Required	Description
job_id	integer	yes	Job ID
parameters	object	no	Parameter overrides as a key/value map

jobs_cancel_run

Parameter	Type	Required	Description
run_id	integer	yes	Run ID

jobs_repair_run

Re-run failed tasks in a multi-task job run.

Parameter	Type	Required	Description
run_id	integer	yes	Run ID of a failed multi-task job
rerun_tasks	array	no	Specific task keys to re-run. If omitted, re-runs all failed tasks.

---

Cluster Tools

clusters_list

Results are paged (default 20 per page). The response includes next_page_token and prev_page_token when more pages are available.

Parameter	Type	Required	Description
cluster_sources	string	no	Comma-separated filter: UI, API, JOB
cluster_states	string	no	Comma-separated filter: RUNNING, TERMINATED, PENDING, etc.
is_pinned	boolean	no	If true, only return pinned clusters
page_size	integer	no	Clusters per page (default 20)
page_token	string	no	Token from a previous response for next/previous page

clusters_get

Parameter	Type	Required	Description
cluster_id	string	yes	Cluster ID

clusters_get_events

Returns recent events including termination reasons.

Parameter	Type	Required	Description
cluster_id	string	yes	Cluster ID
max_results	integer	no	Max events (default 25)

clusters_start

Start a terminated cluster. Does not wait for it to finish starting.

Parameter	Type	Required	Description
cluster_id	string	yes	Cluster ID

clusters_terminate

Terminate a running cluster. This stops the cluster but does not delete it.

Parameter	Type	Required	Description
cluster_id	string	yes	Cluster ID

---

Pipeline Tools

pipelines_list

Parameter	Type	Required	Description
name	string	no	Filter by pipeline name (substring match)
max_results	integer	no	Max results (default 25)

pipelines_get

Returns target catalog/schema, clusters, libraries, and notifications.

Parameter	Type	Required	Description
pipeline_id	string	yes	Pipeline ID

pipelines_list_events

Returns events including update progress, errors, and data quality metrics.

Parameter	Type	Required	Description
pipeline_id	string	yes	Pipeline ID
max_results	integer	no	Max events (default 25)

pipelines_start

Start a pipeline update. Returns the update_id for tracking.

Parameter	Type	Required	Description
pipeline_id	string	yes	Pipeline ID
full_refresh	boolean	no	Full refresh instead of incremental (default false)

pipelines_stop

Parameter	Type	Required	Description
pipeline_id	string	yes	Pipeline ID

---

Workspace Tools

workspace_list

List objects in a workspace directory. Returns path, object_type (NOTEBOOK, DIRECTORY, FILE, REPO), and language (for notebooks).

Parameter	Type	Required	Description
path	string	yes	Workspace path (e.g. /Users/[e-mail])

workspace_export_notebook

For SOURCE format, returns the notebook content as text. For other formats, returns base64-encoded content.

Parameter	Type	Required	Description
path	string	yes	Notebook path in workspace
format	string	no	Export format: SOURCE (default), HTML, JUPYTER, DBC

---

Files Tools

files_list

List files in a DBFS or Volumes path.

Parameter	Type	Required	Description
path	string	yes	dbfs:/... for DBFS, /Volumes/... for Unity Catalog Volumes

files_read

Read a file from Volumes or DBFS. Returns text content for text files, or a size summary for binary files.

Parameter	Type	Required	Description
path	string	yes	Path to the file
max_bytes	integer	no	Max bytes to read (default 1 MB, cap 10 MB)

---

Secrets Tools

secrets_list_scopes

List all secret scopes in the workspace. Returns a sorted JSON array of scope names.

secrets_list_keys

List secret key names in a scope. Values are never returned.

Parameter	Type	Required	Description
scope	string	yes	Secret scope name

---

Permissions Tools

permissions_get_grants

Get Unity Catalog grants on a securable object. Returns principal and privilege list.

Parameter	Type	Required	Description
securable_type	string	yes	One of: CATALOG, SCHEMA, TABLE, VOLUME, FUNCTION
full_name	string	yes	Full name of the securable

permissions_get_object_permissions

Get access control list for a workspace object. Returns principal and permission levels.

Parameter	Type	Required	Description
object_type	string	yes	One of: clusters, jobs, pipelines, sql/warehouses, etc.
object_id	string	yes	Object ID

---

Security

• Authentication uses the Databricks SDK's standard credential chain. Tokens are never passed directly as tool parameters.
• The token_env_var parameter accepts only the name of an environment variable, not the token value itself. The value is read locally and never leaves the machine.
• SQL execution enforces a max_rows cap of 10,000. File reads enforce a max_bytes cap of 10 MB.
• Secret values are never exposed. The secrets_list_keys tool returns key names only.