Dataset of evals call found in phishing kits. They are used to obfuscate sensible information and/or malicious intent.
The PHP files comes from these dataset : https://github.com/marcoramilli/PhishingKitTracker
This dataset is a subset of the public one, where only code that was confirmed as malicious were selected (exfiltration of victim data was confirmed through dynamic analysis)
10,000 malicious evals found in a large dataset of confirmed phishing kits. This dataset comes from a private source, access to this part will be granted on demand, contact the repo owner.