Find and explore data

Preview

This product is subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the Service Specific Terms, and the Additional Terms for Generative AI Preview Products. Pre-GA products are available "as is" and might have limited support. For more information, see the launch stage descriptions.

This page describes how to discover your data entities in the Google Cloud Data Agent Kit extension for Visual Studio Code and learn about them. The extension offers multiple entry points to find your data, ranging from a Catalog explorer and search to agentic chat.

Before you begin

1. Install the extension.
2. Configure settings. Enable all the APIs and configure your project and region.
3. Optional: To interact with Cloud SQL for MySQL, Cloud SQL for PostgreSQL, and AlloyDB for PostgreSQL resources in the Data Agent Kit extension for VS Code, you must configure them so that metadata can be ingested into the Knowledge Catalog. For more information, see:
   • Enable the integration of Knowledge Catalog on your Cloud SQL for MySQL instance
   • Enable the integration of Knowledge Catalog on your Cloud SQL for PostgreSQL instance
   • Enable the Knowledge Catalog integration for your AlloyDB cluster

Explore the Catalog

The Catalog explorer provides a structured view of your data entities. Navigate the Catalog by completing these steps:

1. In the IDE activity bar, click the Google Cloud Data Agent Kit icon.
2. In the Google Cloud Data Agent Kit menu, expand the Catalog section. The explorer displays these data sources in a tree structure:
   • Data entities in BigQuery and BigLake, and AlloyDB, Cloud SQL, and Spanner entities in your current project.
   • BigQuery public datasets.
   • Other projects in your organization that you have access to. Click Load more to see them in the explorer.
3. To navigate the hierarchy, click a node to expand it. For example, when you click a project ID, the explorer displays data sources organized by product name.
4. Click a product name to explore your data entities for that product and continue expanding to view views and tables.
5. Click a data entity to view its details in a new editor tab.
6. Right-click a resource name to see a list of quick actions, such as Query or Load in Spark DataFrame.

Use agentic discovery

To find datasets using natural language, use your agent Chat panel:

1. If the Chat panel isn't displayed, open the command palette by pressing Cmd/Ctrl+Shift+P, then search for and run Open Chat.
2. In the Chat panel, enter a natural language prompt, such as "find the dataset that has customer orders." The agent displays a Thinking state while it searches relevant tables.
3. If the agent prompts you to run gcloud CLI commands to help with the search, click Run for each command.
4. After the agent displays the list of relevant datasets found in your Google Cloud project, search for and explore them in the Catalog or Universal Search.

Use Universal Search

Initiate a universal search from the command palette or the activity bar.

Shortcut

1. Press Cmd/Ctrl+Shift+P to open the command palette.
2. Search for and select Search for Dataset.
3. Enter your search query in the quick picker and press Enter to launch Universal Search in a new editor tab.

Note: Search for Dataset returns data entities, including instances, databases, tables, views, BigQuery datasets, and schemas.

Activity bar

1. In the IDE activity bar, click the Google Cloud Data Agent Kit icon.
2. In the Google Cloud Data Agent Kit menu, expand the Catalog section.
3. Click Universal Search. Universal Search opens in a new editor tab.
4. Enter your search query and press Enter to view the search results.

Filtering

In the Universal Search editor, use the floating Filters panel to narrow results by Scope, Systems, Project, Type, or Location.

View details

After a data entity is identified, view its details in a new editor tab.

• From search results, click an entity to open details for it in a new tab.
• From the Catalog explorer, click a dataset, such as a table name, to launch its details view.

Details tabs

The details editor is organized into vertical tabs to help you understand your data's context and quality.

The tabs that are displayed depend on the type of data entity that you are viewing. For example, the details view of a BigQuery dataset includes the Details and Relationships tabs, while details for a Spanner table include the Schema and Details tabs.

Tab	Functionality
Preview	View a sample of the dataset data.
Schema	View column names, metadata types, descriptions, and other details of the schema for the dataset.
Details	View properties of the data entity.
Insights	Access AI-generated sample queries that help you jumpstart exploring datasets. For more information, see Access data insights in Knowledge Catalog.
Lineage	View a visual map of the origins and descendants of the dataset, along with changes or transformations that were applied. For more information, see About data lineage.
Relationships	View a graph of relationships with the dataset to understand how it relates to other datasets.
Data profile	View statistical summaries of the data. For more information, see Create and use data profile scans.
Data quality	View data quality metrics such as completeness, uniqueness, or freshness for the selected dataset. For more information, see Auto data quality overview.

What's next

• Organizing BigQuery resources
• Introduction to BigLake metastore