Pause / Resume

run.pause(reason=None) asks a run to stop cooperatively; run.resume() lifts it. Neither blocks until the run actually stops or starts - see Lifecycle & Control for what a safe point is and the full state table this page assumes.

run = await deck.runs.start("Approval", {"order_id": "A-1003"})

await run.pause("operator stepped away")
await run.resume()

The reason travels into the log on the run's own control.requested event, so anything reading run.events() later sees why it stopped, not just that it did.

Checking before you call

run.can reads the status this handle last saw - pause, resume, cancel, no answer (an interrupt's shape is per-workflow, not a flag a generic UI can render). It is for a button's enabled state or a branch that would rather not raise:

if run.can.pause:
    await run.pause("operator stepped away")

can is a snapshot from the last time this handle read status, not a live guarantee: the run can end between checking it and acting on it. pause()/resume() are the authoritative answer instead - they raise RunStateError when the state genuinely refuses, and return quietly only when there is nothing left to do (already paused, already over). See Lifecycle & Control for exactly which state accepts which call.