Why I built an AI repair loop that stops after one fix on purpose
Noumenon AI
A few weeks ago I watched my AI coding agent successfully fix a bug. The tests passed. The patch looked clean. The agent reported success. I shipped it. Three hours later I noticed the agent had also "improved" four unrelated files in the same session. One of those improvements quietly removed a guard clause that I had spent a full evening writing two months earlier. There was no test covering that guard clause because it was the kind of thing you write defensively — the bug it prevents doesn't