Lessons from my overly-introspective, self-improving coding agent

I'm curious what to do if agent self-improvement goes a wrong way. How do you evaluate the agent quality? Which way do you return to the last good branch? How to automate this process?