This repository is a result of the experience gained from the 'Objective For The We v3' - "how to use LLMs to detect macOS malware" training.
Here is demonstrated an automated workflow of logging data -> pre-processing that data -> analysing this data with an LLM in order to detect malware (particularly for macOS).
The files held in the data* folders are placeholders, the scripts should generate their own files given the chance to run locally on macOS (although this hasn't been tested thoroughly).
Due to the fact that github likely doesn't run workflows on macOS, the fetch_data.py file doesn't change the placeholder file in data/data.json, hence the output in results.csv recording macOS data, when theoretically it shouldn't.