Dec 24, 2022Ravie LakshmananSoftware Security / Supply Chain

PyPI Packages

Threat actors have published yet another round of malicious
packages to Python Package Index (PyPI) with the goal of delivering
information-stealing malware on compromised developer machines.

Interestingly, while the malware goes by a variety of names like
ANGEL Stealer, Celestial Stealer, Fade Stealer, Leaf $tealer, PURE
Stealer, Satan Stealer, and @skid Stealer, cybersecurity company
Phylum found them all to be copies of W4SP Stealer[1].

W4SP Stealer primarily functions to siphon user data, including
credentials, cryptocurrency wallets, Discord tokens, and other
files of interest. It’s created and published by an actor who goes
by the aliases BillyV3, BillyTheGoat, and billythegoat356.

“For some reason, each deployment appears to have simply tried
to do a find/replace of the W4SP references in exchange for some
other seemingly arbitrary name,” the researchers said[2]
in a report published earlier this week.

CyberSecurity

The 16 rogue modules are as follows: modulesecurity,
informmodule, chazz, randomtime, proxygeneratorbil, easycordey,
easycordeyy, tomproxies, sys-ej, py4sync, infosys, sysuptoer,
nowsys, upamonkws, captchaboy, and proxybooster.

The campaign distributing W4SP Stealer gained traction[3]
around October 2022, although indications are that it may have
started as far back as August 25, 2022. Since then dozens of
additional bogus packages[4] containing W4SP Stealer[5]
have been published on PyPI by the persistent threat actors.

The latest iteration of the activity, for what it’s worth, makes
no obvious to hide its nefarious intentions, except in the case of
chazz, which leverages the package to download obfuscated Leaf
$tealer malware hosted on the klgrth[.]io paste service.

It’s worth noting that previous versions of the attack chains
have also been spotted fetching next-stage Python code directly
from a public GitHub repository that then drops the credential
stealer.

The surge in new copycat variants dovetails with GitHub’s
takedown of the repository that held the original W4SP Stealer
source code, indicating that cybercriminals likely not affiliated
with the operation are also weaponizing the malware to attack PyPI
users.

“Open-source ecosystems such as PyPI, NPM, and the like are huge
easy targets for these kinds of actors to try and deploy this kind
of malware on,” the researchers said. Their attempts will only
become more frequent, more persistent, and most sophisticated.”

The software supply chain security firm, which kept tabs[6]
on the threat actor’s Discord channel, further noted that a
previously flagged package under the name of pystyle[7]
was trojanized by BillyTheGoat to distribute the stealer.

The module has not only racked by thousands of
downloads
[8] each month, but also
started off as an innocuous utility in September 2021 to help users
style console output. The malicious modifications were introduced
in versions 2.1 and 2.2 released on October 28, 2022.

These two versions, which were live on PyPI for about an hour
before they were pulled, are alleged to have gotten 400 downloads,
BillyTheGoat told Phylum in an “unsolicited correspondence.”

“Just because a package is benign today and has shown a history
of being benign for years does not mean it will remain this way,”
the researchers cautioned[9]. “Threat actors have
shown tremendous patience in building legitimate packages, only to
poison them with malware after they have become sufficiently
popular.”

Found this article interesting? Follow us on Twitter [10] and LinkedIn[11] to read more exclusive
content we post.

References

  1. ^
    W4SP
    Stealer
    (thehackernews.com)
  2. ^
    said
    (blog.phylum.io)
  3. ^
    gained
    traction
    (thehackernews.com)
  4. ^
    additional bogus packages
    (thehackernews.com)
  5. ^
    W4SP
    Stealer
    (blog.phylum.io)
  6. ^
    kept
    tabs
    (blog.phylum.io)
  7. ^
    pystyle
    (github.com)
  8. ^
    thousands of downloads
    (pypistats.org)
  9. ^
    cautioned
    (blog.phylum.io)
  10. ^
    Twitter 
    (twitter.com)
  11. ^
    LinkedIn
    (www.linkedin.com)

Read more