Set a twice-daily check for hard-to-find coffee beans

A Dot used old chat context to watch for rare U.S. coffee beans and returned Gesha choices with prices worked out per cup.

Done with Dots· Intermediate

What they did
A ChatGPT Pro user named their Dot 豆哥 and asked it to keep an eye on rare coffee beans in the U.S. The Dot pulled in context from earlier chats, then put itself on a twice-a-day schedule. It kept running on its own cloud computer even after the laptop was closed.
How it went
When it reported back, it surfaced Gesha options and worked the cost out per cup. The piece treated it as one of the stronger early examples because it reused prior context, scheduled follow-up checks, and handled the pricing math.
Tips
Use a clear product type, location, and budget so the agent knows what counts as a match. Ask it to stop and ask before any purchase or checkout step.

Try it yourself with Dots

Track [product type] in [country or region] twice a day at about [time 1] and [time 2]. Use any helpful context from my earlier chats to narrow the search. Look for options from [sites, sellers, or stores] and calculate the cost in [unit, such as per cup, per serving, or per item]. Send me a short update only when you find something relevant, with the item name, seller, price, and link. Before any purchase, checkout, or other irreversible action, pause and ask for my approval first.

Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.

Get 3 agent uses like this every morning

With the day's top AI news. Free, and one click to unsubscribe.