This demo shows how a visual editor could work, using sliders and widgets to tweak config parameters that control a scatter plot of the Iris dataset. The visualization renders sepal length against sepal width, with points colored by species from a configurable color scale. The code uses d3-rosetta’s `unidirectionalDataFlow` and `createStateField` to manage state, along with D3’s `scaleLinear`, `extent`, and `csv` for data loading and scaling. It also listens for `postMessage` events to update the config dynamically.