Skip to main content
Public Preview The extension’s Data Hub panel shows all catalogs, schemas, and tables registered in your Wherobots Organization. Browse what’s available, then use the built-in MCP server to create notebooks that query your data, all without leaving your editor.
Organization Availability: Available to all Wherobots Organization Editions.

​
Before you start

​
Browse the Data Hub

The Data Hub panel is in the Wherobots extension sidebar. It displays a tree of all catalogs registered in your Organization, including any data sources you’ve connected and Wherobots’ built-in curated datasets.
  1. Open the Wherobots extension sidebar by clicking the Wherobots icon in the Activity Bar of VS Code or your Code OSS-based editor.
  2. Expand the Data Hub panel.
  3. Navigate the tree structure:
    • Catalogs are at the top level, labeled as managed or foreign.
    • Expand a catalog to see its schemas.
    • Expand a schema to see its tables.
Use the refresh button at the top of the Data Hub panel to see newly added catalogs or tables after completing an integration.
To add new data source connections, view table metadata, or delete connections, use the Wherobots Data Hub in the Wherobots Cloud web console. For more information, see Wherobots Data Hub.

​
Create a notebook from your data

Once you can see your data in the Data Hub, use the built-in MCP server through your editor’s AI assistant to create notebooks that query it.
The Wherobots MCP Server requires a Professional or Enterprise Organization Edition. Community Edition Organizations do not have access to the Wherobots MCP server.
1

Start a workspace

If you don’t already have a running workspace, create one by running Wherobots: Create Workspace from the Command Palette. For more information, see Workspaces & Usage.
2

Identify the data you want to work with

Browse the Data Hub in the sidebar to find the catalog, schema, and table you want to query. Note the fully qualified name, for example, wherobots_open_data.overture_maps_foundation.buildings_building.
3

Open your AI assistant in Agent mode

Open your editor’s AI assistant (e.g. GitHub Copilot Chat in VS Code) and select Agent mode.
For best results, use an advanced AI model such as the latest Claude Opus, Gemini Pro, or GPT Codex.
4

Prompt the AI assistant to create a notebook

Ask the assistant to build a notebook using the data you identified. For example:
Create a notebook that loads the buildings table from wherobots_open_data.overture_maps_foundation.buildings_building and calculates the total building footprint area per city.
The MCP server handles catalog discovery and query execution behind the scenes. Allow the Wherobots MCP server when prompted.
5

Run the notebook

Open the generated .ipynb file and run the cells against your workspace. For detailed instructions, see Run a local notebook against Wherobots compute.

​
Iterate on your analysis

One of the key benefits of having the Data Hub in your editor is the ability to quickly discover and switch between datasets as your analysis evolves. Rather than starting over when your requirements change, you can pivot to a different dataset with a single prompt. For example, suppose you start an analysis using the Microsoft Buildings dataset, but then realize the Overture Maps buildings data has better coverage for your region of interest. Instead of rewriting your notebook, you can:
  1. Browse the Data Hub to find the Overture Maps tables under wherobots_pro_data.overture_maps_foundation.
  2. Ask the AI assistant to switch:
    The Microsoft Buildings data doesn’t have the coverage I need. Switch to using the Overture Maps buildings dataset from wherobots_pro_data.overture_maps_foundation instead, and re-run the analysis.
The MCP server discovers the new schema and the assistant adjusts your notebook accordingly — updating table references, adapting column names, and re-running cells. This iterative workflow lets you:
  • Explore alternatives: Try different datasets for the same analysis without context-switching.
  • Refine incrementally: Add filters, join additional tables, or change aggregations with follow-up prompts.
  • Validate quickly: Run each iteration against live Wherobots compute to see results immediately.

​
Next steps

AI-Assisted Notebook Development

Learn more about creating and running notebooks with AI assistance.

Workspaces & Usage

Create, start, stop, and destroy workspaces.

Wherobots Data Hub

Manage catalogs and connections in the Wherobots Cloud web console.