Which best captures the key idea of "Prompt libraries"?

AHandle bursts without failures BLLMs are models trained on vast text that predict the most likely next words... CHave people score outputs; it's the gold standard for subjective tasks DMaintain reusable, approved prompts across a team