When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works.

Official Python repository used to distribute cryptomining malware

Obfuscated malware code

Security research firm Sonatype, which focuses on software supply chain management security, has identified six different Python packages containing malware on the official Python software repository PyPI. The malicious packages all contained instructions in their setup.py setup files that would download and install cryptomining malware onto systems that install the packages.

The packages themselves were downloaded almost 5,000 times in the last few months, and were posted by a user named nedog123. The packages in question are:

  • maratlib: 2,371
  • maratlib1: 379
  • matplatlib-plus: 913
  • mllearnlib: 305
  • mplatlib: 318
  • learninglib: 626

Some of them are clearly meant as "typosquats", named 1 character off or very similarly to popular machine learning packages on PyPI such as "mplatlib" instead of the official "matplotlib". This kind of supply-chain security attack is making companies very nervous right now although this one does not seem to do anything more than steal some system resources.

Sonatype details out exactly how they were able to detect the cryptominer, even though the packages were heavily obfuscated and called other packages from github. The blog post is very interesting if malware research is up your alley, and even though it reads a bit like an ad for their Release Integrity software, it is an effective one.

This kind of malware should not affect most users that might be running advanced antivirus protection, these were machine learning packages targeted at researchers with expensive high performance linux machines.

Source: Sonatype Blog via Ars Technica

Report a problem with article
Next Article

Google is now allowing developers to enroll in 15% Play Store commissions

John McAfee delivering a speech
Previous Article

John McAfee, 75, found dead in apparent suicide in Barcelona prison

Join the conversation!

Login or Sign Up to read and post a comment.

1 Comment - Add comment