It works well because they had tests, they could compare before and after. Likelihood of issues especially if you transpile to a safer language should be minimal.
I store all prompts as docs, but I doubt I can rerun the whole prompt history with a better model and get sizable return for the cost, personally.
EDIT: I'm referring to a small codebase not articles tho
Would like to see something more comprehensive built on zfs and freebsd jails. Namely snapshot/checkpoint before each prompt, quick undo for changes made by agent, auto delete old snapshots etc