Databricks

Tool to retrieve information about a specific provider in the Databricks Marketplace with visible listings. Use when you need to get provider details including contact information, description, and metadata.

Untrusted display-only summary

Tool to create a new listing in Databricks Marketplace for data providers. Use when you need to publish data products, datasets, models, or notebooks to the marketplace. Requires a listing object with summary information (name and listing_type). For free and instantly available data products, a share must be included…

Untrusted display-only summary

Tool to retrieve a specific marketplace provider listing by its identifier. Use when you need to get detailed information about a published or draft listing including metadata, configuration, and assets.

Untrusted display-only summary

Tool to create a marketplace consumer installation for Databricks Marketplace listings. Use when you need to install data products, datasets, notebooks, models, or other marketplace offerings into a workspace. Requires acceptance of consumer terms and the listing ID to proceed with installation.

Untrusted display-only summary

Tool to uninstall a Databricks Marketplace installation. Use when you need to remove an installed data product from your workspace. When an installation is deleted, the shared catalog is removed from the workspace. Requires CREATE CATALOG and USE PROVIDER permissions on the Unity Catalog metastore, or metastore admin…

Untrusted display-only summary

Tool to update marketplace consumer installation fields and rotate tokens for marketplace listings. Use when you need to modify installation attributes or refresh access credentials. The token will be rotated if the rotate_token flag is true.

Untrusted display-only summary

Retrieve multiple published marketplace listings by their IDs in a single API call. Use this action when you need to fetch details for multiple listings efficiently instead of making individual GET requests. The action accepts up to 50 listing IDs per request and returns complete listing information including summarie…

Untrusted display-only summary

Tool to retrieve personalization requests for a specific marketplace listing. Use when you need to check the status of customization or commercial transaction requests for a listing. Each consumer can make at most one personalization request per listing.

Untrusted display-only summary

Retrieve multiple marketplace provider details in a single batch API call. Returns information about Databricks Marketplace providers that have at least one publicly visible listing. Use this tool when you need to get details for multiple providers efficiently (up to 50 providers per request). Provider information inc…

Untrusted display-only summary

Tool to create a provider analytics dashboard for monitoring Databricks Marketplace listing metrics. Use when you need to establish analytics tracking for listing views, requests, installs, conversion rates, and consumer information. Requires Marketplace admin role and system tables to be enabled in the metastore.

Untrusted display-only summary

Tool to retrieve provider analytics dashboard information for monitoring consumer usage metrics. Use when you need to access the dashboard ID to view marketplace listing performance including views, requests, installs, and conversion rates.

Untrusted display-only summary

Tool to retrieve the latest logical version of the provider analytics dashboard template. Use when you need to get the current dashboard template version for monitoring consumer usage metrics including listing views, requests, and installs.

Untrusted display-only summary

Tool to remove the association between a marketplace exchange and a listing. Use when you need to disassociate an exchange from a provider listing. This removes the listing from the private exchange, and it will no longer be shared with the curated set of customers in that exchange.

Untrusted display-only summary

Tool to create a new MLflow experiment for tracking machine learning runs and models. Use when you need to organize and track ML experiments within Databricks. Returns RESOURCE_ALREADY_EXISTS error if an experiment with the same name already exists.

Untrusted display-only summary

Tool to create a new logged model in MLflow that ties together model metadata, parameters, metrics, and artifacts. Use when you need to create a LoggedModel object as part of the unified 'log + register' workflow introduced in MLflow 2.8. LoggedModel objects persist throughout a model's lifecycle and provide a central…

Untrusted display-only summary

Tool to create a new MLflow run within an experiment for tracking machine learning execution. Use when starting a new ML training run, experiment execution, or data pipeline that needs parameter and metric tracking. Returns the created run with a unique run_id for subsequent metric and parameter logging.

Untrusted display-only summary

Tool to delete an MLflow experiment and associated metadata, runs, metrics, params, and tags. Use when you need to remove an experiment from Databricks. If the experiment uses FileStore, artifacts associated with the experiment are also deleted.

Untrusted display-only summary

Tool to delete a logged model from MLflow tracking. Use when you need to permanently remove a LoggedModel from the tracking server. The deletion is permanent and cannot be undone. LoggedModels track a model's lifecycle across different training and evaluation runs.

Untrusted display-only summary

Tool to delete a tag from a logged model in MLflow. Use when you need to remove metadata from a LoggedModel object. This operation is irreversible and permanently removes the tag from the logged model. Part of MLflow 3's logged model management capabilities.

Untrusted display-only summary

Tool to mark an MLflow run for deletion in ML experiments. Use when you need to remove a specific run from Databricks. This is a soft delete operation - the run is marked for deletion rather than immediately removed and can be restored unless permanently deleted.

Untrusted display-only summary

Tool to bulk delete runs in an ML experiment created before a specified timestamp. Use when you need to clean up old experiment runs. Only runs created prior to or at the specified timestamp are deleted. The maximum number of runs that can be deleted in one operation is 10000.

Untrusted display-only summary

Tool to delete a tag from an MLflow experiment run. Use when you need to remove run metadata. This operation is irreversible and permanently removes the tag from the run.

Untrusted display-only summary

Tool to finalize a logged model in MLflow by updating its status to READY or FAILED. Use when custom model preparation logic is complete and you need to mark the model as ready for use or indicate that upload failed. This is part of the experimental logged models feature introduced in MLflow 2.8+.

Untrusted display-only summary

Tool to retrieve MLflow experiment metadata by name. Use when you need to get experiment details using the experiment name. Returns deleted experiments but prefers active ones if both exist with the same name. Throws RESOURCE_DOES_NOT_EXIST if no matching experiment exists.

Untrusted display-only summary

Tool to retrieve metadata for an MLflow experiment by ID. Use when you need to get experiment details including name, artifact location, lifecycle stage, and tags. Works on both active and deleted experiments.

Untrusted display-only summary

Tool to fetch logged model metadata by unique ID. Use when you need to retrieve a LoggedModel object representing a model logged to an MLflow Experiment. Returns comprehensive model information including metrics, parameters, tags, and artifact details.

Untrusted display-only summary

Tool to retrieve available permission levels for a Databricks ML experiment. Use when you need to understand what permission levels can be assigned to users, groups, or service principals for a specific experiment. Returns permission levels (CAN_READ, CAN_EDIT, CAN_MANAGE) with their descriptions.

Untrusted display-only summary

Tool to retrieve permissions for an MLflow experiment. Use when you need to check who has access to an experiment and their permission levels. Note that notebook experiments inherit permissions from their corresponding notebook, while workspace experiments have independent permissions.

Untrusted display-only summary

Tool to retrieve complete information about a specific MLflow run including metadata, metrics, parameters, tags, inputs, and outputs. Use when you need to get details of a run by its run_id. Returns the most recent metric values when multiple metrics with the same key exist.

Untrusted display-only summary

Tool to log a batch of metrics, parameters, and tags for an MLflow run in a single request. Use when you need to efficiently log multiple metrics, params, or tags simultaneously. Items within each type are processed sequentially in the order specified. The combined total of all items across metrics, params, and tags c…

Untrusted display-only summary