Skip to content

hemanthy1/AI-Firewall

Repository files navigation

The URL Scanner

The purpose of our project is to develop a Web Application Security Service for the AI Club and any other person that wishes to use.

In the journey of creating this security framework, we saw the importance of the URLs for WAFs, and so we decided to focus our efforts in it.

From our studies of bad HTTP requests to server-side with common attacks such as LFIs, RFIs, and XSS, we noticed that the URLs comprised the main way of identifying the same. This also happened for client-side attacks, related to phishing and spam. So, we decided to focus on their identification and categorization.

Github Contributors


Y0uk1tsun3

felixliang50779

satyabyreddy

pprahlada

hemanthy1

Who We Are

Felipe Marques Allevato | Computer Science | Sophomore | Cybersecurity Lover
Hemanth Yalamanchili | Computer Science | Junior | Learning Machine Learning
Felix | Computer Science | Junior | Aspiring Algorithm Engineer
Pramod Prahlada | Computer Engineering | Junior | Interested in Machine Learning and AI
Satya Byreddy | Computer Science | Junior | Machine Learning Enthusiast

How to set up the App

1. First, clone the repository

git clone https://github.com/MSU-AI/AI-Firewall

Then, remember to have all the requirements installed, for this run the code bellow.

python3 -m pip install -r requirements.txt

2. Prepare your dataset or use the pre-built model

Take a look on the notebooks. The data preparation one will help you create your own dataset and the ML WAF one will help you understand the theory and create your own model!

If you just want to run the application, unpack the zip file and use the pre-generated pkl file :).

3. Running

Check in the utils.py for the port chosen, after that just run the program with flask.

flask --app URL_scanner run

If it does not work, use python to run it.

python3 URL_scanner.py

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published