A collection of spooky scripts that demonstrate the potential of Large Language Models (LLMs) to support CySec tasks.
This repository originated from an initial set of scripts that I wrote while attending the "Generative AI: Boost Your Cybersecurity Career" course by IBM.
Blog post: https://tsumarios.github.io/blog/2024/07/13/genai4cysec/
This repository includes the following scripts:
- ram_reserver.c - A C script designed to allocate a large portion of RAM to get the system under memory pressure (DoS).
- cookie_manipulator.py - A Python script to demonstrate Anti-Forensics manipulation of HTTP cookies.
- docx_terminator.py - A Python script that removes DOCX files.
- keystroke_logger.py - A basic Python keystroke logger to understand the mechanics of keylogging.
- local_port_blocker.py - A Python script to block specified local ports, calling iptables.
- spyware.py - A simple example of Python spyware to illustrate common techniques used in malicious software.
- spyware_detector.sh - A shell script to detect and remove spyware by scanning for known signatures and behaviours.
All the scripts were created with the help of ChatGPT as a demonstration of its support for cybersecurity tasks.
This repository also includes a Jupyter notebook that provides a practical demonstration of how Machine Learning (ML) techniques can be applied for cybersecurity purposes. In particular, the notebook features an example to distinguish between anti-forensics techniques and legitimate privacy practices.
The notebook includes various classification algorithms such as K-Nearest Neighbors, Naive Bayes, Logistic Regression, Support Vector Machines, Random Forest, XGBoost, and Neural Networks. Futhermore, it provides an introduction to Adversarial Machine Learning with an example based on Fast Gradient Sign Method.
As the icing on the cake, the notebook was created with the help of ChatGPT, too.
- Email: marioraciti@pm.me
- LinkedIn: linkedin.com/in/marioraciti
- Twitter: twitter.com/tsumarios
If you want to support me, I would be grateful ❤️
Enjoy!