What do you want in an agent-to-human handoff?
OpenClaw users discuss what a good handoff looks like when an agent hands a task back: what was decided, what was tried and failed, and exactly what's waiting on the human. A bare 'done, over to you' kills context.