schedule-followup

name	schedule-followup
description	Schedule a future check-in that creates a WorkItem for you or another agent at a specific time (one-shot or bounded recurring).
version	1.0.0
category	followup
skillType	claude-skill
tags	["followup","schedule","cron","workitem"]
execution	{"type":"script","script":{"file":"execute.sh","interpreter":"bash","timeoutMs":15000}}

schedule-followup

Schedule a future check-in that creates a WorkItem for you or another agent. Use this when you need to come back to something at a specific time — most commonly to chase a response, verify an async result, or re-prompt an agent that went quiet.

When to use this vs other scheduling skills

Need	Use
"Wake me up in 30 min to check X" (self)	schedule-followup --in-minutes 30 --title "Check X"
"Check back on another agent at 9am"	schedule-followup --fire-at 2026-04-24T09:00:00Z --target agent-session --title "..."
"Hourly for up to 3 tries, then give up"	schedule-followup --cron "0 * * * *" --max-fires 3
"Whenever Rex goes idle, check progress"	watch-for-event (event-based, not this skill)
Simple self-reminder with no team scope	schedule-check (older skill)

Before you schedule — check you aren't already scheduled

Schedules live in two stores: follow-up triggers (this skill) and cron tasks the orchestrator sets for you. Check both before creating, or you will stack a duplicate that fires N× reports:

• list-my-followups — your follow-up triggers
• list-my-crons — recurring crons targeting you (the orchestrator's)

Do not treat "cron list shows 0" alone as "no schedule exists" — verify with both skills, and if still unsure, ask the orchestrator instead of re-creating (#621).

Key invariants

• Every followup is team-scoped. It shows up under list-my-followups and can be cancelled by name via cancel-followup.
• Every followup has a finite lifetime: either maxFires exhausts it, or maxIdleFires (default 3) auto-cancels it after consecutive unproductive fires.
• If you forget to cancel a done followup, it will still terminate itself. That's by design — we refuse to let followups live forever.

Examples

# One-shot self-reminder 30 minutes from now
bash execute.sh --in-minutes 30 --title "Verify Rex delivered the draft"

# One-shot absolute time, targeting Ella
bash execute.sh --fire-at 2026-04-24T09:00:00Z \
  --target crewly-marketing-ella-member-1 \
  --title "Re-prompt Rex if no response yet"

# Recurring with cap — will auto-stop after 3 tries even if you forget to cancel
bash execute.sh --cron "0 * * * *" --max-fires 3 \
  --title "Poll Rex hourly (give up after 3)"

Output

On success, returns the created Trigger object (JSON with id, status, nextFireAt, etc). Keep the id or name if you may need to cancel later.

Cancelling

• By id: cancel-followup --id <trigger-id>
• By name: cancel-followup --name "followup:abc123"
• Or let maxFires/maxIdleFires do it automatically.