Roughly how much does it cost to run to scrape a page? I see from the code this is basically an OpenAI API wrapper but you make no mention of that anywhere on your landing page/documentation, nor any mention of which LLMs this is capable of working with.
Also, an idea is to offer a "record" and "replay" mode. Let the LLM run through the instructions, find the selectors, record and save them. Then you can run through again without using the LLM, replaying the interaction log, until the workflow breaks, then re-generate the "interaction log" or whatever.
This is a great call-out. It's something currently in our roadmap
Re: cost for execution. This really depends on the page, but currently it costs between 5 cents and 20 cents per page to execute (today).
We have an improvement planned to help it "remember" or "cache" actions it's done in the past so it can just replay them and bring the cost down to near zero.
Re: LLMs it's capable of working with, currently it's only GPT-4V. I'll get this updated soon!
Based on #2, it seems like they only use the LLM when the page changes. I had a prototype of this sort of system working and it was surprisingly fault tolerant.
If you want to build it yourself, you could try using https://browserbase.com/. We offer managed headless browsers work everywhere, every-time. It costs $0.10 per browser session/hour (billed minutely). Feel free to shoot me an email if you want access! paul@browserbase.com
Also, an idea is to offer a "record" and "replay" mode. Let the LLM run through the instructions, find the selectors, record and save them. Then you can run through again without using the LLM, replaying the interaction log, until the workflow breaks, then re-generate the "interaction log" or whatever.