Inspiration
While talking to ACV at the career fair, they told us that a main concern they have with their auction site is having web scrapers re-upload their listings elsewhere. This made us start brainstorming ways to help prevent that. This project is more relevant than ever today with the rise of AI agents that can scrape websites very quickly.
What it does
Our demonstration site takes the information it stores in it's database, then it runs it through a randomly generated character mapping, which scrambles the letters of the alphabet. It then gives this obfuscated data to the HTML file, along with a custom generated font so that the data is readable for humans on the website, but appears as jargon to any bots scraping the HTML.
How we built it
The demonstration website was written in Java with Springboot, and our font generation is done in Python. The python code is executed on our server using Graalpy. For our demonstration of it's effectiveness, we are using Crawl4AI, to scrape the site.
Challenges we ran into
Finding out how to modify fonts was a big challenge, because there is a big lack of information online regarding that.
Accomplishments that we're proud of
Creating custom font files to unscramble our obfuscated HTML
What we learned
We learned about how multiple programming languages can work together, and gained more experience working with Springboot and Github.
What's next for Scrambl
Find more creative solutions, and make the concept more scalable.
