Feed
Discussion
I really like this approach. The “ask first, build second” rule sounds simple, but it addresses one of the easiest traps to fall into with AI coding: letting the model quietly make decisions that you never actually made.
The part about paying for a few minutes of questions instead of a much bigger rewrite later really resonated with me. 😅 I’ve seen AI produce perfectly reasonable code based on an assumption that was completely wrong for the actual project.
I also like that you point out this isn’t about a particular /megaplan command. It’s really a workflow habit—make the AI expose its assumptions, answer the important questions, then let it build.
And the reminder that this shouldn’t be used for every tiny task is important too. Forcing a full interview before fixing a typo would probably be worse than the problem itself. 😂
Simple idea, but definitely one I’d like to incorporate into my own AI-assisted development workflow.
Spec first agents only pay off when the spec is the thing you review. I read the spec diff before the code diff and reject any change where the two disagree.
Grunz_ai
asking first fixes the planning half, but whatever it learns has to survive the whole session too. building grunz (coding agent) we had long runs where the original ask got summarized into something vague like "working on the site" and the agent restarted the plan from scratch like it was a new task. so i'd keep the answers to those questions somewhere the agent rereads, not just in chat. does /megaplan save its output anywhere or does it live in the conversation?