Databricks

Delete a secret. Args: - scope: Scope name - key: Secret key Returns: Confirmation of deletion.

List ACLs for a secret scope. Args: - scope: Scope name Returns: List of ACLs.

Get an MLflow experiment by name. Args: - experimentName: Experiment name/path Returns: Experiment metadata.

List all schemas in a catalog. Args: - catalogName: Catalog name Returns: List of schemas.

Get details of a specific cluster. Args: - clusterId: Cluster ID Returns: Cluster configuration and status.

List all clusters in the workspace. Returns: List of clusters with their configuration and status.

Get details of a model version. Args: - name: Model name - version: Version number Returns: Model version details.

Create a new cluster. Args: - clusterName: Name for the cluster - sparkVersion: Spark version (e.g., "13.3.x-scala2.12") - nodeTypeId: Node type ID (e.g., "i3.xlarge") - numWorkers: Number of worker nodes (for fixed-size clusters) - minWorkers: (Optional) Minimum workers for autoscaling - maxWorkers: (Optional) Maximum workers for autoscaling - autoterminationMinutes: (Optional) Auto-terminate after idle minutes Returns: Created cluster ID.

Start a terminated cluster. Args: - clusterId: Cluster ID to start Returns: Confirmation that the cluster is starting.

Restart a running cluster. Args: - clusterId: Cluster ID to restart Returns: Confirmation that the cluster is restarting.

Terminate a running cluster. The cluster configuration is preserved and can be restarted later. Args: - clusterId: Cluster ID to terminate Returns: Confirmation that the cluster is terminating.

Permanently delete a cluster. This removes the cluster completely. Use terminate_cluster to just stop it. Args: - clusterId: Cluster ID to delete Returns: Confirmation of deletion.

List recent events for a cluster. Args: - clusterId: Cluster ID - limit: (Optional) Maximum events to return (default: 50) Returns: List of cluster events.

Pin a cluster to ensure it appears in the cluster list. Args: - clusterId: Cluster ID to pin Returns: Confirmation of pinning.

Unpin a cluster. Args: - clusterId: Cluster ID to unpin Returns: Confirmation of unpinning.

List files and directories in DBFS. Args: - path: DBFS path (e.g., "/mnt/data" or "dbfs:/mnt/data") Returns: List of files and directories with metadata.

Get status/metadata of a DBFS path. Args: - path: DBFS path Returns: File or directory metadata.

Create a directory in DBFS. Creates parent directories as needed. Args: - path: Directory path to create Returns: Confirmation of creation.

Delete a file or directory in DBFS. Args: - path: Path to delete - recursive: (Optional) Delete recursively if directory (default: false) Returns: Confirmation of deletion.

Read contents of a file in DBFS. For files larger than 1MB, use offset and length for chunked reading. Args: - path: File path to read - offset: (Optional) Byte offset to start reading (default: 0) - length: (Optional) Number of bytes to read (default: 1MB, max: 1MB) Returns: Base64-encoded file content and bytes read.

Write content to a file in DBFS. For files larger than 1MB, use the streaming API (create, add_block, close). Args: - path: Destination file path - contents: Base64-encoded content - overwrite: (Optional) Overwrite existing file (default: false) Returns: Confirmation of write.

Move a file or directory in DBFS. Args: - sourcePath: Source path - destinationPath: Destination path Returns: Confirmation of move.

List all instance pools. Returns: List of instance pools.

Get details of a specific instance pool. Args: - instancePoolId: Instance pool ID Returns: Instance pool details.

Create a new instance pool. Args: - instancePoolName: Pool name - nodeTypeId: Node type ID - minIdleInstances: (Optional) Minimum idle instances - maxCapacity: (Optional) Maximum capacity - idleInstanceAutoterminationMinutes: (Optional) Auto-termination minutes Returns: Created instance pool ID.

Edit an existing instance pool. Args: - instancePoolId: Instance pool ID - instancePoolName: New pool name - nodeTypeId: Node type ID - minIdleInstances: (Optional) Minimum idle instances - maxCapacity: (Optional) Maximum capacity Returns: Confirmation of update.

Delete an instance pool. Args: - instancePoolId: Instance pool ID Returns: Confirmation of deletion.

List all jobs in the workspace. Args: - limit: (Optional) Maximum number of jobs to return (default: 20) - offset: (Optional) Offset for pagination - name: (Optional) Filter jobs by name (substring match) Returns: List of jobs with their settings and metadata.

Get details of a specific job. Args: - jobId: Job ID Returns: Job configuration and metadata.

Create a new job. Args: - name: Job name - tasks: Array of task definitions (JSON) - schedule: (Optional) Cron schedule expression - timezone: (Optional) Timezone for schedule (default: UTC) Returns: Created job ID.

Update an existing job. Args: - jobId: Job ID to update - settings: New job settings as JSON string Returns: Confirmation of update.

Delete a job. Args: - jobId: Job ID to delete Returns: Confirmation of deletion.

Trigger a job run immediately. Args: - jobId: Job ID to run - parameters: (Optional) Runtime parameters as JSON string Returns: Run ID of the triggered run.

List job runs. Args: - jobId: (Optional) Filter by job ID - activeOnly: (Optional) Only return active runs - completedOnly: (Optional) Only return completed runs - limit: (Optional) Maximum runs to return (default: 20) - offset: (Optional) Pagination offset Returns: List of job runs with status.

Get details of a specific job run. Args: - runId: Run ID Returns: Run details including status and tasks.

Get the output of a job run. Args: - runId: Run ID Returns: Run output including notebook output if applicable.

Cancel a running job run. Args: - runId: Run ID to cancel Returns: Confirmation of cancellation.

Cancel all active runs of a job. Args: - jobId: Job ID Returns: Confirmation of cancellation.

List MLflow experiments. Args: - maxResults: (Optional) Maximum experiments to return - pageToken: (Optional) Pagination token Returns: List of experiments.

Get details of an MLflow experiment. Args: - experimentId: Experiment ID Returns: Experiment metadata.

Create a new MLflow experiment. Args: - name: Experiment name (workspace path) - artifactLocation: (Optional) Location to store artifacts Returns: Created experiment ID.

Delete an MLflow experiment. Args: - experimentId: Experiment ID Returns: Confirmation of deletion.

Restore a deleted MLflow experiment. Args: - experimentId: Experiment ID Returns: Confirmation of restoration.

Search for MLflow runs. Args: - experimentIds: Array of experiment IDs to search - filter: (Optional) Filter string (e.g., "metrics.accuracy > 0.9") - maxResults: (Optional) Maximum results to return - orderBy: (Optional) Order by fields (e.g., ["metrics.accuracy DESC"]) - pageToken: (Optional) Pagination token Returns: List of matching runs.

Create a new MLflow run. Args: - experimentId: Experiment ID - runName: (Optional) Run name - tags: (Optional) Tags as JSON array [{key, value}] Returns: Created run details.

Update an MLflow run status. Args: - runId: Run ID - status: New status (RUNNING, SCHEDULED, FINISHED, FAILED, KILLED) - endTime: (Optional) End timestamp in milliseconds Returns: Confirmation of update.

Delete an MLflow run. Args: - runId: Run ID Returns: Confirmation of deletion.

Log a metric to an MLflow run. Args: - runId: Run ID - key: Metric name - value: Metric value - timestamp: (Optional) Timestamp in milliseconds - step: (Optional) Step number Returns: Confirmation of logging.

Log a parameter to an MLflow run. Args: - runId: Run ID - key: Parameter name - value: Parameter value Returns: Confirmation of logging.

Set a tag on an MLflow run. Args: - runId: Run ID - key: Tag name - value: Tag value Returns: Confirmation of setting.

List registered MLflow models. Args: - maxResults: (Optional) Maximum models to return - pageToken: (Optional) Pagination token Returns: List of registered models.

Get details of a registered model. Args: - name: Model name Returns: Model metadata including versions.

Create a new registered model. Args: - name: Model name - description: (Optional) Model description - tags: (Optional) Tags as JSON array [{key, value}] Returns: Created model details.

Delete a registered model. Args: - name: Model name Returns: Confirmation of deletion.

List versions of a registered model. Args: - name: Model name Returns: List of model versions.

Delete a model version. Args: - name: Model name - version: Version number Returns: Confirmation of deletion.

List Delta Live Tables pipelines. Args: - maxResults: (Optional) Maximum pipelines to return - pageToken: (Optional) Pagination token - filter: (Optional) Filter expression Returns: List of pipelines.

Get details of a specific pipeline. Args: - pipelineId: Pipeline ID Returns: Pipeline details including status and configuration.

Create a new Delta Live Tables pipeline. Args: - name: Pipeline name - spec: Pipeline specification as JSON string Returns: Created pipeline ID.

Update an existing pipeline. Args: - pipelineId: Pipeline ID - spec: Updated pipeline specification as JSON string Returns: Confirmation of update.

Delete a pipeline. Args: - pipelineId: Pipeline ID Returns: Confirmation of deletion.

Start a pipeline update. Args: - pipelineId: Pipeline ID - fullRefresh: (Optional) Perform a full refresh (default: false) Returns: Update ID.

Stop a running pipeline. Args: - pipelineId: Pipeline ID Returns: Confirmation of stop.

List Git repos in the workspace. Args: - pathPrefix: (Optional) Filter by path prefix - pageToken: (Optional) Pagination token Returns: List of repos.

Get details of a specific repo. Args: - repoId: Repo ID Returns: Repo details including branch info.

Create a new Git repo in the workspace. Args: - url: Git repository URL - provider: Git provider (e.g., gitHub, bitbucketCloud, gitLab) - path: (Optional) Workspace path for the repo Returns: Created repo details.

Update a repo (switch branch or tag). Args: - repoId: Repo ID - branch: (Optional) Branch to checkout - tag: (Optional) Tag to checkout Returns: Updated repo details.

Delete a repo from the workspace. Args: - repoId: Repo ID Returns: Confirmation of deletion.

List Git credentials for the current user. Returns: List of Git credentials.

Create a new Git credential. Args: - provider: Git provider (e.g., gitHub, bitbucketCloud, gitLab) - username: Git username - personalAccessToken: Personal access token Returns: Created credential details.

Delete a Git credential. Args: - credentialId: Credential ID Returns: Confirmation of deletion.

List all secret scopes. Returns: List of secret scopes.

Create a new secret scope. Args: - scope: Scope name Returns: Confirmation of creation.

Delete a secret scope. This also deletes all secrets and ACLs in the scope. Args: - scope: Scope name Returns: Confirmation of deletion.

List secrets in a scope. Note: This lists secret metadata only, not the actual values. Args: - scope: Scope name Returns: List of secret keys.

Create or update a secret. Args: - scope: Scope name - key: Secret key - value: Secret value (will be stored securely) Returns: Confirmation of storage.

Get ACL for a principal on a scope. Args: - scope: Scope name - principal: Principal name (user or group) Returns: ACL details.

Set ACL for a principal on a scope. Args: - scope: Scope name - principal: Principal name (user or group) - permission: Permission level (READ, WRITE, or MANAGE) Returns: Confirmation of setting.

Delete ACL for a principal on a scope. Args: - scope: Scope name - principal: Principal name (user or group) Returns: Confirmation of deletion.

Execute a SQL statement on a Databricks SQL warehouse. The statement will be executed and results returned inline if they complete within the timeout. For long-running queries, use databricks_get_sql_status to poll for completion. Args: - warehouseId: SQL warehouse ID to execute on - statement: SQL statement to execute - catalog: (Optional) Default catalog for the statement - schema: (Optional) Default schema for the statement - waitTimeout: (Optional) Time to wait for results (e.g., "50s", default: "50s") - format: (Optional) Result format: JSON_ARRAY, CSV, or ARROW_STREAM Returns: Statement ID and status. If completed, includes result data.

Get the status and results of a SQL statement. Use this to poll for completion of long-running queries or to retrieve results. Args: - statementId: Statement ID from executeStatement Returns: Statement status and results if available.

Get a specific chunk of SQL statement results. For large results that are split into multiple chunks, use this to fetch each chunk. Args: - statementId: Statement ID - chunkIndex: Chunk index (0-based) Returns: Result data for the specified chunk.

Cancel a running SQL statement. Args: - statementId: Statement ID to cancel Returns: Confirmation of cancellation.

List all SQL warehouses in the workspace. Returns: List of SQL warehouses with their status and configuration.

Get details of a specific SQL warehouse. Args: - warehouseId: SQL warehouse ID Returns: Warehouse configuration and status.

Start a stopped SQL warehouse. Args: - warehouseId: SQL warehouse ID to start Returns: Confirmation that the warehouse is starting.

Stop a running SQL warehouse. Args: - warehouseId: SQL warehouse ID to stop Returns: Confirmation that the warehouse is stopping.

List all personal access tokens for the current user. Returns: List of token metadata (not the actual token values).

Get details of a specific function. Args: - fullName: Full function name (catalog.schema.function) Returns: Function metadata.

Create a new personal access token. Args: - comment: (Optional) Token description - lifetimeSeconds: (Optional) Token lifetime in seconds (default: 90 days) Returns: Token value (shown only once) and metadata.

Revoke a personal access token. Args: - tokenId: Token ID to revoke Returns: Confirmation of revocation.

List all catalogs in Unity Catalog. Returns: List of catalogs with metadata.

Get details of a specific catalog. Args: - name: Catalog name Returns: Catalog metadata and properties.

Create a new catalog in Unity Catalog. Args: - name: Catalog name - comment: (Optional) Catalog description Returns: Created catalog details.

Delete a catalog. Args: - name: Catalog name - force: (Optional) Force delete even if not empty (default: false) Returns: Confirmation of deletion.

Get details of a specific schema. Args: - fullName: Full schema name (catalog.schema) Returns: Schema metadata.

Create a new schema in a catalog. Args: - catalogName: Catalog name - name: Schema name - comment: (Optional) Schema description Returns: Created schema details.

Delete a schema. Args: - fullName: Full schema name (catalog.schema) Returns: Confirmation of deletion.

List all tables in a schema. Args: - catalogName: Catalog name - schemaName: Schema name Returns: List of tables.

Get details of a specific table. Args: - fullName: Full table name (catalog.schema.table) Returns: Table metadata including columns.

Delete a table. Args: - fullName: Full table name (catalog.schema.table) Returns: Confirmation of deletion.

List all volumes in a schema. Args: - catalogName: Catalog name - schemaName: Schema name Returns: List of volumes.

Get details of a specific volume. Args: - fullName: Full volume name (catalog.schema.volume) Returns: Volume metadata.

Create a new volume. Args: - catalogName: Catalog name - schemaName: Schema name - name: Volume name - volumeType: Volume type (MANAGED or EXTERNAL) - storageLocation: (Optional) Storage location for external volumes - comment: (Optional) Volume description Returns: Created volume details.

Delete a volume. Args: - fullName: Full volume name (catalog.schema.volume) Returns: Confirmation of deletion.

List all functions in a schema. Args: - catalogName: Catalog name - schemaName: Schema name Returns: List of functions.

List contents of a workspace directory. Args: - path: Workspace path (e.g., "/Users/user@example.com") Returns: List of workspace objects (notebooks, folders, etc.).

Get status/metadata of a workspace object. Args: - path: Workspace path to the object Returns: Object metadata including type, language, and timestamps.

Create a directory in the workspace. Creates parent directories as needed. Args: - path: Directory path to create Returns: Confirmation of creation.

Delete a workspace object (notebook, folder, etc.). Args: - path: Path to delete - recursive: (Optional) Delete recursively if folder (default: false) Returns: Confirmation of deletion.

Import a notebook into the workspace. Args: - path: Destination path for the notebook - content: Base64-encoded content - format: Import format (SOURCE, HTML, JUPYTER, DBC, R_MARKDOWN) - language: (Optional) Notebook language (PYTHON, SCALA, SQL, R) - required for SOURCE format - overwrite: (Optional) Overwrite existing file (default: false) Returns: Confirmation of import.

Export a notebook from the workspace. Args: - path: Path to the notebook - format: (Optional) Export format (SOURCE, HTML, JUPYTER, DBC, R_MARKDOWN) - default: SOURCE Returns: Base64-encoded notebook content.