Privacy Policy Checker

Ship #1

@Christie on Privacy Policy Checker · 13 days ago

This is a simple policy checker that allows you to enter a URL and check its policies.

As a security-conscious person, I always found myself really frustrated that the 10000 legal policies every company has are impossible to a) find and b) parse, because they’re written in legalese.

I built this tool to help me to find and check the policies of websites we use in normal English, using AI to summarize and explain them.

I learnt how to use Python to scrape websites!
It was super interesting to see how you can automate something that would otherwise involve a lot of manual searching and copying. One thing I found particularly interesting was considering the ethical side of things. Just because information is publicly accessible doesn’t necessarily mean it’s okay to collect and use it however you want, but also scraping a page is basically the same thing as visiting it, so I don’t really know
what my opinion is here.
After spending a long time working mostly with JavaScript, it was also nice to return to Python and refresh my memory!
That doesn’t mean it was easy though – debugging my code was sometimes super difficult as I had lots of moving parts :)

Getting to call AI models through an API was also pretty cool! It was interesting to see how you can integrate AI into your own code and build tools around it, rather than just interacting with AI through a chat interface.

And finally, a huge thank you to Hack Club for providing free AI access and making it possible to experiment with these tools! It’s awesome to have the opportunity to learn, build, and explore new ideas without worrying about API costs.
Also thanks to Claude for helping me deploy my project!

ALSO ALSO IF YOU’RE READING THIS, THANK YOU SOO MUCH!! I REALLY WANT A RETRO COMPUTER MAGAZINE SO PLEASSEEE BE NICE TO ME :»>

6 devlogs
13h
12.15x multiplier

Try project → See source code →

Open comments for this post

@Christie on Privacy Policy Checker · 14 days ago

3h 43m 40s logged

Sorry about the bullet points, short on time.py

Made SECRET_KEY mandatory: app now raises RuntimeError at startup if it’s missing, instead of silently running with None
Added rate limiting via Flask-Limiter, returns a friendly 429 message when exceeded, added Flask_Limiter to requirements
Wrapped the AI content-processing call in a try/except so a failure there shows an error message instead of crashing the request
-Turned off debug mode for production
added a “Back” button
Fixed HTML typos
Added a little bit of CSS!
Deployed with Nest!!!!
Yayyy I’m done!!

Open comments for this post

@Christie on Privacy Policy Checker · about 1 month ago

4h 43m 5s logged

I finished the logic for my app, putting together all the previous functions with Flask to make a functioning web app. Now all I have to do is add CSS!

Open comments for this post

@Christie on Privacy Policy Checker · about 2 months ago

2h 35m 33s logged

I added a bunch of stuff to my app!
To explain simply:
-> I scrape the content off all the links using Beautiful Soup
-> I push this content into AI to be summarised using Hack Club’s AI API
-> Summarised terms and conditions are returned!
This means I can work on my frontend now yayyyy!!!

Open comments for this post

@Christie on Privacy Policy Checker · 3 months ago

28m 13s logged

Added scanning of robots.txt and sitemap.xml
CHECKPOINTTT
I finished building the first stage of the web app by coding the second and third ways way discover legal document links – scanning robots.txt and sitemap.xml using Requests, Python and BeautifulSoup, and filtering them against a list of keywords such as “privacy”, “terms”, and “cookies” and removing duplicates. Now I have three ways of finding links, and can move onto actually reading what’s inside them!
Yay :)

Open comments for this post

@Christie on Privacy Policy Checker · 3 months ago

48m 55s logged

Adding Scraping Robots.txt Functionality
To get a more comprehensive link check, I made it so that my link searcher also scraped robots.txt of a domain, where more potential links may be lurking (hehehe).
This is important because it will help me find more terms to scan, making my tool better!
This was done in a similar process to the scraping footers function. However, I used response/request to get the robots.txt file instead of scanning as page with BeautifulSoup.
Just as before, I scanned the links and checked for duplicates before adding to the main array.
Yay!
Below, you can see it in action (I combined the two functions).

Open comments for this post

@Christie on Privacy Policy Checker · 3 months ago

21m 27s logged

I focused on building the first stage of the web app by coding the first way discover legal document links – scanning page footers using Requests and BeautifulSoup, and filtering them against a list of keywords such as “privacy”, “terms”, and “cookies” and removing duplicates.
I also began implementing a second discovery method using robots.txt, which can now be loaded.
This didn’t take me 19 minutes – over 2 hours, in fact – but I only discovered Stardance later.