Pompous and full of air really sums it up well. As such I usually have to completely ignore and rework generated README / code comments. Is this just an elabor

Pompous and full of air really sums it up well. As such I usually have to completely ignore and rework generated README / code comments. Is this just an elaborate scheme to increase in/out token churn or are we hitting some kind of LLM limitations I wonder?