I didn't know that (and BTW Ruby-Toolbox is a great source of inspiration)
I think the challenge with that is that there's no standardization of categories. Both GitHub and npm have the concept of "topics/keywords" that developers can choose. Problem is that every maintainer chooses slightly different keywords, so it becomes ineffective as a tool for discovery.
We still haven't fully figured out what the optimal solution might be like.
I think it would be a good start, but would still require some human vetting.
We played with some simple NLP solutions (trying to run TF-IDF on the keywords/description/readme of packages to determine the category), the results weren't great honestly, so we opted of curating everything manually for now.