Kanaries/pygwalker ? reverse-engineered prompt

I want to create a Python library that makes exploratory data analysis in Jupyter Notebooks much more intuitive. Instead of writing a bunch of matplotlib or seaborn code, I want to be able to take a pandas dataframe and instantly turn it into a full interactive user interface for building visualizations. It should basically work like a mini Tableau or Power BI, but running directly within the notebook cell.

The main idea is a drag-and-drop interface. I should be able to see all my columns and just drag them to create different charts like bar charts, scatter plots, or line charts. I'd want to be able to quickly change chart types, add multiple measures to compare, and even create facet views to see sub-groups in my data. The goal is to provide a powerful, visual way to explore a dataset without needing to write a single line of visualization code after loading the data.