Skip to main content

ChatterBox

ChatterBox is Ziggy's own UI for searching any of your data.

Source Code​

The source code is freely available should you want to modify and tailor it to your own requirements. It is an React application.

How it works​

ChatterBox requires basic customization. For standard cases, this can be done in an hour or so. However, almost any aspect can be fine-tuned and enhanced so more complex requirements will take longer.

We can perform these customizations for you, or you can do it yourself with the source code.

Here are a few key points.

  • ChatterBox connects with a simple Ziggy Flow to perform the actual search.
  • You can connect to multiple data targets. Each different target has a button in the header area.
  • You use natural language queries to perform any search.
  • The search results are automatically displayed as listings, tables or crosstabs, depending on the nature of the data returned.
  • Aggregation queries can show a pie, line or bar chart.
  • Listings can be customised to show whatever fields you want to see and how each field should be formatted.
  • You can drill down from any record to display some or all fields in a customizable card.
  • The history of all queries and results are saved ot the listing so you can recall them at any time.
  • Your query history can also be shown and managed.

Search Flows​

When ChatterBox users enter a query, ChatterBox passes control to a Ziggy Flow. These Flows are simple and quickly configured.

Most Search Flows will contain an AI Prompt Block and a platform specific search Block.

Data is returned to ChatterBox in a standardised format so ChatterBox can instantly render in the appropriate listing format.

Note that all Ziggy Flows are API callable so you can manage the UI yourself if you prefer.

Want us to configure ChatterBox (or Flows) for you?​

If you want any assistance configuring any aspect of Ziggy, including ChatterBox, please contact us to discuss.