Use poke for reminders, alarms, recurring nudges, deferred agent work,
reply-driven escalation, and reminder side effects such as scripts that
start before delivery or stop when the user replies.
sub-skills/*.md file before building the command.--dry-run for unfamiliar flag combinations.poke --reply; read sub-skills/replies.md first when handling replies.
| If making or handling... | Read first |
|---|---|
| --- | --- |
| Basic one-shot, exact-time, recurring reminders, or deferred tasks | sub-skills/scheduling.md |
| "Keep poking until I reply", escalation, rotating tones, heartbeat-style checks, unconfirmed followups, task intervals | sub-skills/escalation.md |
| Pre/post actions, side-effect scripts, wake lights, or stop-on-reply hooks | sub-skills/stages.md |
| User replies such as done, cancel, snooze, later, or ambiguous inbound messages | sub-skills/replies.md |
| Quiet hours, active hours, DND, urgent delivery, or "do not wake me" behavior | sub-skills/quiet-hours.md |
| Multiple channels, dependency chains, or per-channel visibility | sub-skills/multi-channel.md |
| Listing, showing, cancelling, history, stats, presets, tones, path sets, or state layout | sub-skills/management.md |
| Inferred promises like "I'll check on that tomorrow" without a precise fire time | sub-skills/commitments.md |
| Migrating OpenClaw heartbeat polling into scheduled poke tasks | sub-skills/migration.md |
--remind is for static reminder text.--task is for deferred agent work and needs --agent.--channel and --target identify where delivery and replies happen.--pre-cmd, --post-cmd, --on-pong-command, --if-unconfirmed-command) execute real local programs under your user account at scheduled time — only configure hooks with scripts you wrote or explicitly trust.共 2 个版本