Ten years ago a similar scraper took them a horrendous amount of time and wasn't half as good. With free Claude and ChatGPT, this one grabs images, reads them with OCR and crops the hits.
Result About 5-10 hours over a weekend
“Nothing particularly advanced but it would have taken me a horrendous amount of time without it to be half as good, like it did when I built a similar scraper 10 years ago.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
Used free-tier Claude and ChatGPT to build a scraper for items on a website they follow. It grabs the item images and runs a first pass with local OCR; when keywords match, it uses OpenCV to crop the image for better OCR, then saves the hits for a closer look.
It took about 5-10 hours over a weekend. Nothing advanced, they say, but a similar scraper they built 10 years ago took a horrendous amount of time and wasn’t half as good.
Help me build a script that visits [website section] every [period], downloads the item images and runs OCR on each one. If the text contains any of [keywords], crop the relevant area with OpenCV, run OCR again on the crop, and save the image, the text and the link to a review folder. I'm not a programmer, so explain how to run it.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.