This seems to suffer from what a lot of vibe coded projects suffer from, which is just a whole lot of unnecessary text everywhere. I find one shot outputs always have a lot of jargon, extra "micro copy" and displays the same information multiple times in multiple places. It makes it difficult at a glance to understand what you're looking at.

I'm not sure what in the training set causes the models to do this, even in my own experience I'll be specific about the feature I'm building and the model will add an unnecessary status message for no reason.