Over 170K users hit by poisoned Python package ruse

Over 170K users hit by poisoned Python package ruse

March 25, 2024 at 02:06PM

Over 170,000 users were impacted by a complex attack employing fake Python infrastructure. The attack targeted the Top.gg GitHub organization and other developers, distributing malware-infected Python PyPI packages. This led to data theft from browsers, Discord, and crypto wallets. The attack involved various tactics, including creating clones of popular Python packages and compromising trusted GitHub accounts. The attacker managed to insert the fake package into the GitHub repository. The extent of user impact is unclear, but the attack highlights the challenges in defending open source package managers against such multifaceted attacks.

From the meeting notes, it is clear that an attack targeted the Top.gg GitHub organization and other developers, resulting in the successful exploitation of multiple victims. The attackers used various supply chain attack techniques to distribute malware-infected Python PyPI packages, which led to the theft of data from browsers, Discord, crypto wallets, and files.

The attack involved multiple tactics such as cloning popular Python packages, registering a doppelganger domain for Python packages, and breaking into trusted GitHub community members’ accounts. The attack began in November 2022 when the malicious Python packages were uploaded, but it intensified in February when the doppelganger domain was registered.

The attackers ensured the fake URL strings looked nearly identical to the real URL, allowing them to host the malware-infected packages undetected. Additionally, they replaced instances of the genuine PyPI domain with the doppelganger URL on GitHub to further conceal their activities.

The compromised GitHub repository of a 170,000-strong Discord server suggests that at least thousands or even tens of thousands of users may have been affected by the malware.

The attack highlights the challenges of defending against sophisticated attacks that use multiple vectors to exploit open source package managers like PyPI.

Full Article