Use image scans to spot thrift-store flips
An agent examined thrift-store images, matched items to online listings, and pointed out buys that looked cheap enough to resell.
- What they did
- Wes Bos put together a custom agent that watched a thrift-store browsing session through captured frames. It recognized the items shown, checked what similar pieces were selling for online, and called out the ones priced well below those comparisons.
- How it went
- Bos said the agent performed surprisingly well. It successfully highlighted bargains from the scan instead of just naming objects.
- Tips
- Better results will come from clear close-up shots and a simple comparison target, such as resale marketplaces. Ask for a ranked shortlist so you can focus on the strongest candidates first.
Try it yourself with Gemini
I’ll upload thrift-store photos or short clips from a shopping trip. Identify each item you can make out, estimate a normal resale value by checking [eBay, Etsy, or other marketplaces], and compare that estimate with the store’s asking price. Point out the purchases that appear under market value, then stop and wait for my approval before any step that would involve buying, sending, deleting, or submitting anything. Finish with a ranked list showing the item, asking price, estimated resale price, and your confidence score.
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.
Related agent uses
Let an AI agent place a lunch order
A developer had an AI agent handle lunch ordering and payment, and the meal showed up without any more input.
Let an agent choose and order healthy meals
A developer had an agent pick and buy the healthiest meal options twice a day for two weeks.
Have an agent compare shirts and order one
An agent compared shirt options across six brands, chose a size, found the closest store, and completed the purchase.
Automate weekly store updates from a spreadsheet
A custom agent keeps an online shop current by reading a weekly inventory sheet, revising listings, and asking for approval on unclear items.
Build a no-purchase agent for delivered prices and coupons
Have an agent total the landed cost, test coupons, and check buyer feedback before it reaches checkout.
Use Signal to assemble a shared grocery cart
A self-hosted Signal agent turned past orders into a group-editable grocery list and then filled the store cart, leaving checkout to a person.