​
What is ScraperSensei?

Welcome to ScraperSensei! We’re a service that helps you extract data from websites and automate your workflows right in your browser.

​
Ready to get started?

Quickstart

Learn how to get started with ScraperSensei in minutes

Features

Tips for Prompting

Reference

Learn more about the ScraperSensei API

Glossary

Learn about the terms used in the ScraperSensei documentation

​
FAQ

​
Can ScraperSensei smartly plan the actions according to my one-line goal? Like executing “Tweet ‘hello world’”

ScraperSensei is an automation assistance SDK with a key feature of action stability — ensuring the same actions are performed in each run. To maintain this stability, we encourage you to provide detailed instructions to help the AI understand each step of your task. If you require a ‘goal-to-task’ AI planning tool, you can develop one based on ScraperSensei. Related Docs:

​
Limitations

There are some limitations with ScraperSensei. We are still working on them.
  1. The interaction types are limited to only tap, type, keyboard press, and scroll.
  2. It’s not 100% stable. Even GPT-4o can’t return the right answer all the time. Following the Prompting Tips will help improve stability.
  3. Making AI able to automate your browser accurately is challenging. ScraperSensei needs servers that will receive your OpenAI API key to performs some tasks like OCR, drawing boxes (like microsoft/Omniparser) and dealing with the OpenAI result to make result more accurate. We don’t store your OpenAI API key, and you can revoke access anytime if you want.

​
About the token cost

Here are some typical data with GPT-4o.
TaskResolutionPrompt Tokens / PriceCompletion Tokens / Price
Plan the steps to search on eBay homepage1280x8006,975 / $0.034875150 / $0.00225
Locate the search box on the eBay homepage1280x8008,004 / $0.0400292 / $0.00138
Query the information about the item in the search results1280x80013,403 / $0.06701595 / $0.001425

​
What data is sent to LLM ?

Currently, the contents are:
  1. the key information extracted from the DOM, such as text content, class name, tag name, coordinates;
  2. a screenshot of the page.