Inspiration
Inspired from the workshop, where one of our most used services were mentioned to be very invasive. We wondered what the key points of those extremely long TOS and privacy policy are, and if there is a service that can provide a TL;DR for the these terms.
What it does
It scrapes TOS or privacy policy from a given service name/domain by scraping relevant data. Then, the AI would create a concise analysis of the key points that stands out. There are multiple levels predefined for the users to select, from 'Elementary Schooler' to 'Teacher'.
How we built it
Built with Python for the server, and pure HTML/CSS/JS for the client side.
Challenges we ran into
The web scraper failed occasionally on certain websites. Thus, added a summarization method that uses google search results automatically for fallback options.
Accomplishments that we're proud of
We are proud that our app has fallback options, and on testing environment worked with all cases(full url, incomplete url, just the service name) thanks to that.
What we learned
We have learned that you can't get everything right on the first try. Especially in programming, it can take several attempts to reach the desired result, but given enough time and patience, it will eventually work.
What's next for TosAutomation
Our next step is first revising the FastAPI architecture and make server handle more complex tasks, and make it more easier to be expandable. Then, our second step would be using more modern frontend such as Next.JS, to add more functionality with ease. Continual chat for follow-up questions, better UI elements is currently being considered. Finally, our project ToSAutomation eventually aim to index important data on vector database to implement RAG-based advanced query resolving, and add 'deep analysis' capabilities using multi-step method to find out every single small details and related legal cases of the service's policy.
Log in or sign up for Devpost to join the conversation.