Copy-paste prompt
Every Monday and Thursday at 9am, load open invoices from [Stripe / QuickBooks / sheet]. Group them by how many days overdue. Draft a reminder for each, getting firmer as the bucket gets older, and include the payment link. Log the reminders in [sheet] so nobody gets more than one a week. Check that each invoice is still unpaid before drafting. Don't send anything.
Fill in everything in [square brackets]. Put your approval rules (see the free operator's charter) above it.
What you need
- Invoice list export (Stripe, QuickBooks, sheet)
- Email connector
How the agent does it
- Load open invoices with the due date and amount.
- Bucket them by days overdue: 1-7, 8-30, 31-60, 60+.
- Draft a reminder whose tone fits the bucket: friendly, then firm, then final notice.
- Track the reminder history so nobody gets two in a week.
What you get
An aging table and one draft per overdue invoice, with the payment link included.
| Schedule | Mondays and Thursdays at 9am. |
|---|---|
| Needs your approval | Sending reminders, adding late fees and writing off debts need approval. |
| How to check it worked | Totals reconcile with the invoicing system; no client is reminded twice in 7 days. |
Common mistakes
- Chasing invoices that were already paid. Re-check status right before drafting.
Setting it up in each agent
In Grok Bot
Ask the Bot to run the prompt once and review the result. Then say "Save the process we just used as a skill called 'Unpaid invoice chaser'" and "Create a routine that runs the Unpaid invoice chaser skill mondays and thursdays at 9am". Grok Bot's docs say a Bot can own up to 50 routines, schedules must be at least five minutes apart, and you should use Test run before you enable one.
In Meta Muse
Paste the prompt into Muse and ask it to remember the job and run it on the schedule. Muse asks for approval before outbound actions and purchases, so keep the 'don't send' line in the prompt. Before you rely on it, check that Muse can reach the apps this recipe needs.
In ChatGPT Dots
Give your dot the prompt as an ongoing goal and set the schedule. Dots can work between chats and ask for approval before sensitive actions. OpenAI says background work counts toward your plan's allowance, so match the frequency to what you actually need.
Run it by hand once with harmless inputs before you schedule it. A test run does real work.