All products

Build Plotly charts with AI

Give your agent a table and the question behind it. Get a Plotly figure with labelled axes, details on hover and an export suited to its readers.

To build Plotly charts with AI, tell the agent what decision depends on the figure. A list of weekly delivery times might call for a line to show a trend, or a box plot to show how much the times vary between depots.

The visualization skill keeps shared settings in one place in a Python notebook or script. Change the type size there for a projector, then keep the underlying values, units and hover labels available to someone inspecting the figure at their desk.

What it does

  • Chart the spread of fulfilment times at each depot, including unusually slow orders.
  • Rank stock categories with horizontal bars when their names need extra room.
  • Give anyone hovering over a point its unit, period and series name.
  • Add a creation date so colleagues can identify the chart they saw last week.
  • Move a chart from interactive HTML into an illustrated report as a static image.

Install and use

  1. Install the skill in your agent and make Python and Plotly available. Static image export also needs Kaleido and its compatible browser.
  2. Provide the column names and units. Ask, “Which depots have the widest range of delivery times?”
  3. Check the titles and hover values against one known row, then choose an offline HTML file or a static export.

FAQ

Source

Made by Atlassian. Original skill: visualization, published on GitHub under the Apache-2.0 licence.

View on GitHub