Ask HN: Does Claude Code succeed after being asked "should we give up?" for you?
Almost every time Claude has failed to properly implement something or fix a bug several times, if I ask "should we give up?" it declares it's not giving up and successfully completes the task. Of course, it could very well be that when I resort to this tends to coincide with Claude's progress on a problem, but absent rigorous testing, it's truly uncanny. I can't decide whether to be happy I have a "win" button or exasperated that I "have" to go through the ritual. I think you’re on to something.
Chatbots prioritize continuing the conversation. I have found that they will happily spin in circles till you start to give up.
I suspect what you’re doing is helping it break out of a thinking rut, take a small big picture look and try a last time.
I’ve seen similar break loop tactics work. (Entirely new content window, pause for several hours or overnight and reframe, etc)