Integration

Databricks integration

Databricks unified analytics platform API for managing workspaces, clusters, jobs, notebooks, and SQL warehouses. Execute queries, manage data pipelines, and orchestrate machine learning workflows.

Connect Databricks How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Databricks.Built from the 31 Databricks tools below, on the connection you add once.

Common Databricks use cases

automate cluster management
automate sql warehouse management
automate job management
automate run management
automate secret scope management
automate notebook management
get job programmatically
get run programmatically
list job programmatically
list run programmatically

What you can do with Databricks

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Cluster, 6 operations

create_clusterCreate clusterPOSTCreate a new cluster in the workspace. Configure Spark version, node types, autoscaling, and other settings. The cluster starts automatically after creation.
delete_clusterDelete clusterPOSTTerminate a cluster by its ID. The cluster is removed from the running state but its configuration is preserved for potential restart.
get_clusterGet clusterGETRetrieve detailed information about a specific cluster by its ID. Returns cluster state, Spark version, node types, autoscaling configuration, and runtime details.
list_clustersList clustersGETList all available clusters in the workspace. Returns information about each cluster including its state, configuration, and creator. Use to monitor cluster inventory and find cluster IDs for other operations.
restart_clusterRestart clusterPOSTRestart a running cluster. All running jobs on the cluster will be terminated and the cluster will be restarted.
start_clusterStart clusterPOSTStart a previously terminated cluster. The cluster resumes with its existing configuration.

Sql warehouse, 6 operations

create_sql_warehouseCreate sql warehousePOSTCreate a new SQL warehouse with specified size and configuration. The warehouse starts automatically after creation.
delete_sql_warehouseDelete sql warehouseDELETEDelete a SQL warehouse by its ID. The warehouse is permanently removed and cannot be recovered.
get_sql_warehouseGet sql warehouseGETRetrieve detailed information about a specific SQL warehouse including its state, size, configuration, and capacity.
list_sql_warehousesList sql warehousesGETList all SQL warehouses in the workspace. Returns warehouse configuration, state, and capacity information.
start_sql_warehouseStart sql warehousePOSTStart a stopped SQL warehouse.
stop_sql_warehouseStop sql warehousePOSTStop a running SQL warehouse. Active queries will be terminated.

Job, 5 operations

create_jobCreate jobPOSTCreate a new job in the workspace (Jobs API 2.0, single-task format). Supply exactly one cluster (existing_cluster_id or new_cluster) and exactly one task (notebook_task, spark_jar_task, spark_python_task, spark_submit_task, pipeline_task or run_j...
delete_jobDelete jobPOSTDelete a job by its ID. All run history and settings for the job are permanently removed.
get_jobGet jobGETRetrieve detailed information about a specific job including its settings, schedule, and cluster configuration.
list_jobsList jobsGETList all jobs in the workspace. Returns job settings, creation time, and creator for each job.
update_jobUpdate jobPOSTPartially update an existing job. Top-level fields present in new_settings are replaced; fields you omit keep their current values, and arrays are merged by key. Use fields_to_remove to delete top-level fields.

Run, 3 operations

cancel_runCancel runPOSTCancel an active job run. The run transitions to a TERMINATED state.
get_runGet runGETRetrieve metadata and status of a specific job run including its state, result, start time, and cluster details.
list_runsList runsGETList runs for a specific job or all jobs. Returns run metadata, state, and results. Supports pagination and filtering by active/completed status.

Secret scope, 3 operations

create_secret_scopeCreate secret scopePOSTCreate a new secret scope in the workspace. Secret scopes are used to store and manage secrets such as database credentials and API keys.
delete_secret_scopeDelete secret scopePOSTDelete a secret scope and all its associated secrets from the workspace.
list_secret_scopesList secret scopesGETList all secret scopes available in the workspace.

Notebook, 2 operations

export_notebookExport notebookGETExport a notebook or file from the workspace. Returns the content in the specified format.
import_notebookImport notebookPOSTImport a notebook or file into the workspace. Supports source code, HTML, Jupyter, and DBC formats.

Job now, 1 operation

run_job_nowRun job nowPOSTTrigger an immediate run of a job. Optionally override job parameters for this specific run.

Run output, 1 operation

get_run_outputGet run outputGETRetrieve the output and metadata of a single task run. Returns the notebook output, error information, and logs.

Sql statement, 1 operation

execute_sql_statementExecute sql statementPOSTExecute a SQL statement on a SQL warehouse. Returns results inline for small result sets or provides an external link for large results. Supports parameterized queries.

Sql statement status, 1 operation

get_sql_statement_statusGet sql statement statusGETCheck the status and retrieve results of a previously executed SQL statement. Use for polling async queries.

Workspace item, 1 operation

list_workspace_itemsList workspace itemsGETList the contents of a workspace directory. Returns notebooks, folders, libraries, and other objects at the specified path.

Workspace status, 1 operation

get_workspace_statusGet workspace statusGETGet the status of a workspace object (notebook, directory, library, or file) at the specified path.

Use Databricks from Claude Code, Cursor or Codex

Connect Databricks in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Databricks operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Databricks

Create your workspace
Sign up free. Nothing to install.
Connect Databricks once
Add your Databricks connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

Databricks integration FAQ

What can you do with the Databricks integration?

LiveContext exposes 31 operations on Databricks, covering cluster, sql warehouse, job and run. Common uses include automate cluster management, automate sql warehouse management and automate job management.