How to keep Claude's warnings from bothering you

A follow-up to the auto-click post. Sometimes, in the middle of a long session, Claude Code suddenly decides the conversation looks suspicious and blocks itself. It stops, and it gently suggests that maybe you should switch over to a different model, Opus 4.8, as if that were the fix. I did not want to switch. I wanted to stay on Fable the whole way, fully automatic, without babysitting it.

This is where two small things I already had turned out to help each other.

First, the auto-clicker. Funny enough, Claude wrote that clicker for me in the first place. I had it running for another reason, but it does the job here too. The moment a block dialog pops up, the clicker just dismisses it before I even read it. The interruption never really lands.

Second, one word in the instructions. I tell it to work atomically. Instead of trying to do the whole thing in one big step that trips the block again, it advances one small piece at a time, and each piece is small enough to go through.

Neither of these was built for this. The clicker was for clicking, the atomic instruction was for keeping long tasks on track. Put together, they quietly route around a wall that was not supposed to be routed around. That is the part I keep noticing lately. The useful stuff is rarely the thing you built for the purpose. It is two old things leaning on each other. And credit where it is due, one of them was written by Claude itself.