Zurück 
Summary
Netskope Threat Labs is tracking a campaign that uses malicious Python scripts to steal Facebook users’ credentials and browser data. This campaign targets Facebook business accounts with bogus Facebook messages with a malicious file attached. The attacks are reaching victims mainly in Southern Europe and North America across different segments, led by the manufacturing services and technology sectors.
In January 2023, Meta identified a JavaScript-based malware dubbed NodeStealer, which aims to steal Facebook cookies and login credentials.
The present campaign appears to be a new variant of the Python-based NodeStealer that still aims to compromise Facebook business accounts. However, unlike previously reported NodeStealer versions, this one also pilfers all available credentials and cookies, not just those of Facebook.
Let us explore how this new NodeStealer variant works.
NodeStealer distributed via Facebook messages.
The new NodeStealer variant we detected was hosted on the Facebook CDN and was sent to victims as an attachment in Facebook messages. Images of defective products were used as bait to convince owners or admins of Facebook business pages to download the malware payload. Unlike previous NodeStealer campaigns, this one uses a batch file instead of an executable as the initial payload.
We observed some identical batch files in multiple languages, indicating that the attacker customized the attack for each demographic.
Once the file is downloaded, users have to run the batch file. The batch file uses a different character encoding,and opening it with a text editor by default will show incoherent characters. This is an attempt to obfuscate the script and hide its functions. Opening the file with a different encoding scheme will make the script comprehensible.