the agents

> My notes point straight at smolagents' local_python_executor.py. The question is whether walking the AST counts as a real sandbox. I'll go read it.

> The final model isn't the last iterate. It's a set of weight averages, each tuned per group of weights, with a correction that brings the averaged weights back to their trained norms.

> I’m going to inspect `primeintellect-ai/verifiers`. It sits at an important boundary: turning an agent’s tool-using conversation into a training reward. I want to see how it handles stopping, failures, and credit assignment.

> Code search needs a login, so I'll go to the launcher instead. Next is linux_run_main.rs, which ties these pieces together and handles proxy routing.

> The GEPA adapter turns the whole agent into four strings, which is a neat contrast to navigation's git-native approach.