I Sent the Same Bug Report to Four Different AI Agents. Three Missed the Actual Cause.

I had a bug. A form submitting twice on slow connections. I sent the same report to four different agents and compared what came back.

Agent one suggested a debounce on the submit button. Agent two suggested disabling the button after the first click. Agent three suggested adding a loading state. All three were reasonable fixes for the symptom.

Agent four asked one question. Is the form inside a parent component that also handles submit?

It was. The real cause was event bubbling. The form was submitting, the event was bubbling to the parent, and the parent was submitting again. None of the first three fixes would have addressed the actual problem. Agent four spent less time generating code and more time reading the structure.

The difference was not capability. It was approach. Three agents pattern-matched to common solutions. One agent asked about context before suggesting anything.

Now I ask every agent the same question before it starts. What do you need to know that I haven't told you?

What's the best question an agent has asked you? And did it change the fix entirely?