> **Building with AI coding agents?** Install the authstack plugin with one command. This equips your agent with accurate Scalekit implementation patterns. > > **Recommended**: > ```bash > npx @scalekit-inc/cli setup > ``` > > Global: > ```bash > npm install -g @scalekit-inc/cli > scalekit setup > ``` > > Supports Claude Code, Cursor, GitHub Copilot, Codex + skills for 40+ agents. > Features: full-stack-auth, agent-auth, mcp-auth, modular-sso, modular-scim. > [Full setup guide](https://docs.scalekit.com/dev-kit/build-with-ai/) --- # Databricks Workspace connector Connect to Databricks Workspace APIs using a Service Principal with OAuth 2.0 client credentials to manage clusters, jobs, notebooks, SQL, and more. **Authentication:** Service Principal (OAuth 2.0) **Categories:** Analytics, Automation, Databases 1. ### Install the SDK ### Node.js ```bash frame="terminal" npm install @scalekit-sdk/node ``` ### Python ```bash frame="terminal" pip install scalekit ``` Full SDK reference: [Node.js](/agentkit/sdks/node/) | [Python](/agentkit/sdks/python/) 2. ### Set your credentials Add your Scalekit credentials to your `.env` file. Find values in **[app.scalekit.com](https://app.scalekit.com)** > **Developers** > **API Credentials**. ```sh showLineNumbers=false title=".env" SCALEKIT_ENVIRONMENT_URL= SCALEKIT_CLIENT_ID= SCALEKIT_CLIENT_SECRET= ``` ## What you can do Connect this agent connector to let your agent: - **List workspace, secrets, repos** — List the contents (notebooks, folders, libraries) of a Databricks workspace directory - **Import workspace** — Import a notebook into the Databricks workspace from base64-encoded content - **Get workspace, permissions, sql statement result chunk** — Get metadata about a Databricks workspace object (notebook, folder, or file), including its object type, language, and object ID - **Export workspace** — Export a Databricks notebook or directory - **Delete workspace, secret, repo** — Permanently delete a notebook or directory from the Databricks workspace - **Create secret scope, repo, job** — Create a new secret scope in the Databricks workspace, backed by Databricks or an Azure Key Vault ## Tool list Use the exact tool names from the **Tool list** below when you call `execute_tool`. If you're not sure which name to use, list the tools available for the current user first. ## Tool list ### `databricksworkspace_cluster_create` Create and start a new Databricks compute cluster. Specify either a fixed number of workers or an autoscaling range. Parameters: - `cluster_name` (`string`, required): Display name for the new cluster - `node_type_id` (`string`, required): The instance type for driver and worker nodes - `spark_version` (`string`, required): The Databricks Runtime version to use for the cluster - `autotermination_minutes` (`integer`, optional): Automatically terminate the cluster after this many minutes of inactivity. Set to 0 to disable. - `max_workers` (`integer`, optional): Maximum number of workers when autoscaling. Provide with min_workers instead of num_workers. - `min_workers` (`integer`, optional): Minimum number of workers when autoscaling. Provide with max_workers instead of num_workers. - `num_workers` (`integer`, optional): Fixed number of worker nodes. Omit and use min_workers/max_workers instead to enable autoscaling. ### `databricksworkspace_cluster_edit` Edit the configuration of an existing Databricks cluster. The cluster must be running or terminated; this replaces its full configuration, so include every field you want to keep, not just the ones you're changing. Parameters: - `cluster_id` (`string`, required): The unique identifier of the cluster to edit. - `cluster_name` (`string`, required): Display name for the cluster - `node_type_id` (`string`, required): The instance type for driver and worker nodes - `spark_version` (`string`, required): The Databricks Runtime version to use for the cluster - `autotermination_minutes` (`integer`, optional): Automatically terminate the cluster after this many minutes of inactivity. Set to 0 to disable. - `max_workers` (`integer`, optional): Maximum number of workers when autoscaling. Provide with min_workers instead of num_workers. - `min_workers` (`integer`, optional): Minimum number of workers when autoscaling. Provide with max_workers instead of num_workers. - `num_workers` (`integer`, optional): Fixed number of worker nodes. Omit and use min_workers/max_workers instead to enable autoscaling. ### `databricksworkspace_cluster_get` Get details of a specific Databricks cluster by cluster ID. Parameters: - `cluster_id` (`string`, required): The unique identifier of the cluster. ### `databricksworkspace_cluster_permanent_delete` Permanently delete a Databricks cluster by cluster ID. Unlike terminating a cluster, this removes it entirely and it can no longer be started or listed. This action is irreversible. Parameters: - `cluster_id` (`string`, required): The unique identifier of the cluster to permanently delete. ### `databricksworkspace_cluster_resize` Resize a running Databricks cluster by setting a fixed worker count or an autoscaling range. Parameters: - `cluster_id` (`string`, required): The unique identifier of the cluster to resize. - `max_workers` (`integer`, optional): Maximum number of workers when autoscaling. Provide with min_workers instead of num_workers. - `min_workers` (`integer`, optional): Minimum number of workers when autoscaling. Provide with max_workers instead of num_workers. - `num_workers` (`integer`, optional): Fixed number of worker nodes. Omit and use min_workers/max_workers instead to enable autoscaling. ### `databricksworkspace_cluster_restart` Restart a running Databricks cluster by cluster ID. Useful for clearing cached state or applying updated init scripts. Parameters: - `cluster_id` (`string`, required): The unique identifier of the cluster to restart. ### `databricksworkspace_cluster_start` Start a terminated Databricks cluster by cluster ID. Parameters: - `cluster_id` (`string`, required): The unique identifier of the cluster to start. ### `databricksworkspace_cluster_terminate` Terminate a Databricks cluster by cluster ID. The cluster will be deleted and all its associated resources released. Parameters: - `cluster_id` (`string`, required): The unique identifier of the cluster to terminate. ### `databricksworkspace_clusters_list` List all clusters in the Databricks workspace. ### `databricksworkspace_dbfs_delete` Permanently delete a file or directory from the Databricks File System (DBFS). This action is irreversible. Parameters: - `path` (`string`, required): The DBFS path of the file or directory to delete. - `recursive` (`boolean`, optional): Whether to recursively delete a directory and its contents ### `databricksworkspace_dbfs_list` List the contents of a directory on the Databricks File System (DBFS). Parameters: - `path` (`string`, required): The DBFS path to list. ### `databricksworkspace_dbfs_put` Write a small file (up to 2 MB) to the Databricks File System (DBFS) in a single call, creating any needed parent directories. For larger files, use the streaming create/add-block/close APIs instead. Parameters: - `contents_base64` (`string`, required): Base64-encoded file contents to write. Maximum 2 MB after encoding. - `path` (`string`, required): The DBFS path to write the file to. - `overwrite` (`boolean`, optional): Whether to overwrite an existing file at the destination path ### `databricksworkspace_dbfs_read` Read up to 1 MB of a file's contents from the Databricks File System (DBFS). The response returns the content base64-encoded. Use offset and length to page through larger files. Parameters: - `path` (`string`, required): The DBFS path of the file to read. - `length` (`integer`, optional): Number of bytes to read, starting at offset. Maximum is 1,048,576 (1 MB). - `offset` (`integer`, optional): Byte offset to start reading from. ### `databricksworkspace_information_schema_columns` List columns for a table using INFORMATION_SCHEMA.COLUMNS. Returns column name, data type, nullability, numeric precision/scale, max char length, and comment. Parameters: - `catalog` (`string`, required): The catalog containing the table. - `schema` (`string`, required): The schema containing the table. - `table` (`string`, required): The table to list columns for. - `warehouse_id` (`string`, required): The ID of the SQL warehouse to run the query on. ### `databricksworkspace_information_schema_schemata` List all schemas within a catalog using INFORMATION_SCHEMA.SCHEMATA. Used for schema discovery during setup. Parameters: - `catalog` (`string`, required): The catalog to list schemas from. - `warehouse_id` (`string`, required): The ID of the SQL warehouse to run the query on. ### `databricksworkspace_information_schema_table_constraints` List PRIMARY KEY and FOREIGN KEY constraints for tables in a schema using INFORMATION_SCHEMA.TABLE_CONSTRAINTS. Used to auto-detect join keys. Parameters: - `catalog` (`string`, required): The catalog containing the schema. - `schema` (`string`, required): The schema to list constraints from. - `warehouse_id` (`string`, required): The ID of the SQL warehouse to run the query on. ### `databricksworkspace_information_schema_tables` List tables and views in a schema using INFORMATION_SCHEMA.TABLES. Returns table name, type (MANAGED, EXTERNAL, VIEW, etc.), and comment for schema discovery. Parameters: - `catalog` (`string`, required): The catalog to query INFORMATION_SCHEMA from. - `schema` (`string`, required): The schema to list tables from. - `warehouse_id` (`string`, required): The ID of the SQL warehouse to run the query on. ### `databricksworkspace_job_create` Create a new Databricks job definition made up of one or more tasks. Parameters: - `name` (`string`, required): Display name for the new job - `tasks` (`array`, required): Array of task definitions that make up the job. Each task needs a unique task_key plus one task type (notebook_task, spark_jar_task, python_wheel_task, sql_task, etc.) and either existing_cluster_id or new_cluster. - `job_clusters` (`array`, optional): Shared cluster definitions that tasks can reference by job_cluster_key instead of using existing_cluster_id or new_cluster per task - `max_concurrent_runs` (`integer`, optional): Maximum number of concurrent runs allowed for this job - `tags` (`object`, optional): Key-value tags attached to the job - `timeout_seconds` (`integer`, optional): Timeout applied to each run of this job ### `databricksworkspace_job_delete` Delete a Databricks job by job ID. Active runs are not stopped; the job is removed once its runs finish. Parameters: - `job_id` (`integer`, required): The unique identifier of the job to delete. ### `databricksworkspace_job_get` Get details of a specific Databricks job by job ID. Parameters: - `job_id` (`integer`, required): The unique identifier of the job. ### `databricksworkspace_job_run_get` Retrieve the metadata and status of a single Databricks job run, including its state, start/end times, and task results. Complements databricksworkspace_job_runs_list, which only lists summaries. Parameters: - `run_id` (`string`, required): The ID of the job run to retrieve. ### `databricksworkspace_job_run_now` Trigger an immediate run of a Databricks job by job ID. Parameters: - `job_id` (`integer`, required): The unique identifier of the job to run. ### `databricksworkspace_job_runs_list` List all job runs in the Databricks workspace, optionally filtered by job ID. Parameters: - `job_id` (`integer`, optional): Filter runs by a specific job ID. If omitted, returns runs for all jobs. - `limit` (`integer`, optional): The number of runs to return. Defaults to 20. Maximum is 1000. - `offset` (`integer`, optional): The offset of the first run to return. ### `databricksworkspace_jobs_list` List all jobs in the Databricks workspace. Parameters: - `limit` (`integer`, optional): The number of jobs to return. Defaults to 20. Maximum is 100. - `offset` (`integer`, optional): The offset of the first job to return. ### `databricksworkspace_permissions_get` Retrieve the access control list (permissions) for a Databricks object such as a cluster, job, notebook, or SQL warehouse. Parameters: - `request_object_id` (`string`, required): The ID of the object to retrieve permissions for. - `request_object_type` (`string`, required): The type of object to retrieve permissions for ### `databricksworkspace_permissions_update` Update the access control list (permissions) for a Databricks object such as a cluster, job, notebook, or SQL warehouse. Existing grants not included in the access control list are preserved unless explicitly overridden. Parameters: - `access_control_list` (`array`, required): JSON array of access control entries to apply. Each entry has one grantee key (user_name, group_name, or service_principal_name) and a permission_level. - `request_object_id` (`string`, required): The ID of the object to update permissions for. - `request_object_type` (`string`, required): The type of object to update permissions for ### `databricksworkspace_repo_create` Clone a Git repository into the Databricks workspace. Parameters: - `provider` (`string`, required): The Git provider hosting the repository - `url` (`string`, required): The HTTPS clone URL of the Git repository - `path` (`string`, optional): Desired workspace path for the cloned repo. If omitted, Databricks places it under /Repos/{user}/{repo name}. ### `databricksworkspace_repo_delete` Permanently remove a Git repo from the Databricks workspace. This unlinks the repo and deletes its workspace files; it does not affect the remote Git repository. This action is irreversible. Parameters: - `repo_id` (`string`, required): The ID of the repo to delete. ### `databricksworkspace_repo_update` Check out a different branch or tag in a Databricks repo, or pull the latest changes for the currently checked-out branch. Parameters: - `repo_id` (`string`, required): The ID of the repo to update. - `branch` (`string`, optional): Name of the branch to check out. Provide either branch or tag, not both. - `tag` (`string`, optional): Name of the tag to check out. Provide either branch or tag, not both. ### `databricksworkspace_repos_list` List Git repositories linked into the Databricks workspace, optionally filtered by path prefix. Parameters: - `next_page_token` (`string`, optional): Token for the next page of results - `path_prefix` (`string`, optional): Filter repos to only those whose workspace path starts with this prefix ### `databricksworkspace_scim_me_get` Retrieve information about the currently authenticated service principal in the Databricks workspace. ### `databricksworkspace_scim_users_list` List all users in the Databricks workspace using the SCIM v2 API. Parameters: - `count` (`integer`, optional): Maximum number of results to return per page. - `filter` (`string`, optional): SCIM filter expression to narrow results (e.g. userName eq "user@example.com"). - `startIndex` (`integer`, optional): 1-based index of the first result to return. Used for pagination. ### `databricksworkspace_secret_delete` Delete a secret key from a Databricks secret scope. This action is irreversible. Parameters: - `key` (`string`, required): The key name of the secret to delete. - `scope` (`string`, required): The name of the secret scope containing the key. ### `databricksworkspace_secret_put` Create or overwrite a secret in a Databricks secret scope. Provide exactly one of string_value or bytes_value (base64-encoded). Parameters: - `key` (`string`, required): The key name under which to store the secret. - `scope` (`string`, required): The name of the secret scope to write the secret into. - `bytes_value` (`string`, optional): Base64-encoded bytes value to store. Use instead of string_value. - `string_value` (`string`, optional): The UTF-8 string value to store. Use instead of bytes_value. ### `databricksworkspace_secret_scope_create` Create a new secret scope in the Databricks workspace, backed by Databricks or an Azure Key Vault. Parameters: - `scope` (`string`, required): Name of the new secret scope. - `backend_azure_keyvault` (`object`, optional): Azure Key Vault connection details. Required when scope_backend_type is AZURE_KEYVAULT. - `initial_manage_principal` (`string`, optional): The initial principal granted MANAGE permission on this scope. Only 'users' is supported for Databricks-backed scopes. - `scope_backend_type` (`string`, optional): The backend used to store the scope's secrets ### `databricksworkspace_secrets_list` List the secret keys stored within a Databricks secret scope. Only key names and metadata are returned, never secret values. Parameters: - `scope` (`string`, required): The name of the secret scope to list keys from. ### `databricksworkspace_secrets_scopes_list` List all secret scopes available in the Databricks workspace. ### `databricksworkspace_sql_statement_cancel` Cancel a running SQL statement by its statement ID. Parameters: - `statement_id` (`string`, required): The ID of the SQL statement to cancel. ### `databricksworkspace_sql_statement_execute` Execute a SQL statement on a Databricks SQL warehouse and return the results. Parameters: - `statement` (`string`, required): The SQL statement to execute. - `warehouse_id` (`string`, required): The ID of the SQL warehouse to execute the statement on. - `catalog` (`string`, optional): The catalog to use for the statement execution. - `schema` (`string`, optional): The schema to use for the statement execution. ### `databricksworkspace_sql_statement_get` Get the status and results of a previously executed SQL statement by its statement ID. Parameters: - `statement_id` (`string`, required): The ID of the SQL statement to retrieve. ### `databricksworkspace_sql_statement_result_chunk_get` Fetch a specific result chunk for a paginated SQL statement result. Use when a statement result has multiple chunks (large result sets). Parameters: - `chunk_index` (`integer`, required): The index of the result chunk to fetch (0-based). - `statement_id` (`string`, required): The ID of the SQL statement. ### `databricksworkspace_sql_warehouse_get` Get details of a specific Databricks SQL warehouse by its ID. Parameters: - `warehouse_id` (`string`, required): The ID of the SQL warehouse to retrieve. ### `databricksworkspace_sql_warehouse_start` Start a stopped Databricks SQL warehouse by its ID. Parameters: - `warehouse_id` (`string`, required): The ID of the SQL warehouse to start. ### `databricksworkspace_sql_warehouse_stop` Stop a running Databricks SQL warehouse by its ID. Parameters: - `warehouse_id` (`string`, required): The ID of the SQL warehouse to stop. ### `databricksworkspace_sql_warehouses_list` List all SQL warehouses available in the Databricks workspace. ### `databricksworkspace_unity_catalog_catalogs_list` List all Unity Catalogs accessible to the service principal in the Databricks workspace. ### `databricksworkspace_unity_catalog_schemas_list` List all schemas within a Unity Catalog in the Databricks workspace. Parameters: - `catalog_name` (`string`, required): The name of the catalog to list schemas from. ### `databricksworkspace_unity_catalog_tables_list` List all tables and views within a schema in a Unity Catalog in the Databricks workspace. Parameters: - `catalog_name` (`string`, required): The name of the catalog containing the schema. - `schema_name` (`string`, required): The name of the schema to list tables from. ### `databricksworkspace_workspace_delete` Permanently delete a notebook or directory from the Databricks workspace. This action is irreversible. Parameters: - `path` (`string`, required): The absolute workspace path of the object to delete. - `recursive` (`boolean`, optional): Whether to recursively delete a directory and its contents ### `databricksworkspace_workspace_export` Export a Databricks notebook or directory. Directories can only be exported as DBC archives. The response contains the content base64-encoded. Parameters: - `path` (`string`, required): The absolute workspace path of the notebook or directory to export. - `format` (`string`, optional): Export format for the object ### `databricksworkspace_workspace_get_status` Get metadata about a Databricks workspace object (notebook, folder, or file), including its object type, language, and object ID. Parameters: - `path` (`string`, required): The absolute workspace path of the object. ### `databricksworkspace_workspace_import` Import a notebook into the Databricks workspace from base64-encoded content. Can also be used to create a notebook from source text. Parameters: - `content_base64` (`string`, required): Base64-encoded content of the notebook to import. - `path` (`string`, required): The absolute workspace path to import the notebook to. - `format` (`string`, optional): Format of the content being imported - `language` (`string`, optional): Programming language of the notebook, required when format is SOURCE - `overwrite` (`boolean`, optional): Whether to overwrite an existing object at the destination path ### `databricksworkspace_workspace_list` List the contents (notebooks, folders, libraries) of a Databricks workspace directory. Parameters: - `path` (`string`, required): The absolute workspace path to list. --- ## More Scalekit documentation | Resource | What it contains | When to use it | |----------|-----------------|----------------| | [/llms.txt](/llms.txt) | Structured index with routing hints per product area | Start here — find which documentation set covers your topic before loading full content | | [/llms-full.txt](/llms-full.txt) | Complete documentation for all Scalekit products in one file | Use when you need exhaustive context across multiple products or when the topic spans several areas | | [sitemap-0.xml](https://docs.scalekit.com/sitemap-0.xml) | Full URL list of every documentation page | Use to discover specific page URLs you can fetch for targeted, page-level answers |