Uncover Hidden Threats in Seconds with Next-Gen Malware Analysis

Malware documents employing geofencing have become a significant threat to cybersecurity. These malicious files often employ location-based triggers, making detection and mitigation a challenging task. However, Adaptive Threat Analysis stands out from traditional approaches by offering the capability to accurately emulate and falsify the expected geolocation values, effectively neutralizing the tactics employed by malware, thus enhancing our ability to protect against such threats.

In the sample provided below, we can observe a geofencing malware attempting to execute exclusively within a specific country. However, our innovative solution successfully bypasses this restriction, as previously mentioned, by emulating the desired geolocation values, demonstrating our superior capability in countering such geofencing-based threats.

By rendering suspicious websites and subjecting them to our advanced machine learning engine we're capable of identifying nearly 300 brands. In the example provided below, you can witness a Russian website masquerading as a computer gaming company known as Steam. Our solution excels in comparing the site's content to the genuine URL, swiftly identifying such fraudulent attempts to safeguard your digital assets and personal information.

The offline URL detector ML model provides a new layer of defense by effectively detecting suspicious URLs, offering a robust means to identify and mitigate threats posed by malicious links. It leverages a dataset containing hundreds of thousands of URLs, meticulously labeled as either no threat or malicious by reputable vendors, to assess the feasibility of accurately detecting suspicious URLs through machine learning techniques.

It is important to note that this feature is particularly useful in air-gapped environments where online reputation lookups are not available.

The sample below reveals a malware that was packed using the UPX packing technique. Despite its attempt to evade detection and defenses, our analysis successfully unpacked the payload, exposing its true identity as a Dridex Trojan. We were able to uncover the malware configuration, shedding light on the malicious intent behind this threat, extracting valuable IOCs.

Employing Similarity Search functionality, sandbox has detected a file remarkably resembling a known malware. Notably, this file had been previously marked as non-malicious, revealing the potential for false negatives in our security assessments. This discovery empowers us to specifically target and rectify these overlooked threats.

It is important to highlight that Similarity Search is highly valuable for threat research and hunting, as it can help uncover samples from the same malware family or campaign, providing additional IOCs or relevant information about specific threat activities.

Our disassembling engine revealed intriguing findings within the target sample. Surprisingly, this sample monitors the system time using the uncommon <rdtsc> instruction and accesses an internal, undocumented structure in Windows, commonly used for different malicious tricks. These unusual actions raise questions about its purpose and underscore the need for further investigation to assess potential risks to the system.

The sample under examination was built using .NET framework. While we refrain from displaying the actual CIL, our decompilation process extracts and presents noteworthy information, including strings, registry artifacts, and API calls.

Besides that, we parse the .NET metadata to identify .NET-specific functions and resources. This process allows to extract detailed information about the assembly, such as methods, classes, and embedded resources, which is critical for analyzing the behavior and structure of .NET applications.

Many application exploits bring their final payload in raw binary format (shellcode), which might be an obstacle when parsing the payload. With our shellcode emulation we are able to discover and analyse the behaviour of the final payload, in this example for a widely leveraged Office vulnerability in the equation editor. Hence opening the door to gathering the relevant IOCs.

Obfuscated VBA macros present a significant challenge to deliver a reasonable response time of active threats. This unclear code makes the analysis and understanding of threats a high complex task that demands a lot of time and efforts. Our cutting-edge VBA emulation technology is able to overcome these challenges and provides a comprehensive analysis of obfuscated VBA macro together with clear insights into its functionality in seconds.

The analyzed sample is an Excel document with highly obfuscated VBA code that drops and runs a .NET DLL file, together with a LNK file in charge of continuing the malware execution chain. After VBA emulation, MetaDefender Sandbox identifies launched processes and the main deobfuscating function, automatically extracts obfuscated strings and saves dropped files (previously hardcoded and encrypted in the VBA code). This rapidly show the main purpose of the malware and give us the possibility of a further analysis of this threat.

Using Windows Task Scheduler to execute malicious payloads at a later time is a stealthy technique to evade sandbox environments seen in recent threats. It exploits the delay in execution to effectively bypass the short analysis window typical of sandboxes.

The following sample is an obfuscated VBScript that downloads the malicious payload and creates a scheduled task to run it 67 minutes later. Traditional sandboxes maintain the execution for only a few minutes and the malicious behavior would be never exposed. In the other hand, our VBScript emulator is able to detect and overcomes this evasion technique (T1497), adapting the execution environment to continue with further analysis, and getting the full report in 12 seconds.

NET Reflection is a powerful feature provided by the .NET framework that allows programs to inspect and manipulate a .NET file structure and behavior at runtime. It enables the examination of assemblies, modules, and types, as well as the ability to dynamically create instances of types, invoke methods, and access fields and properties.

Malware can use reflection to dynamically load and execute code from assemblies that are not referenced at compile time, allowing to fetch additional payloads from remote servers (or hidden in the current file) and execute them without writing them to disk, reducing the risk of detection.

In this case, we can see how the analysed VBScript loads and runs a .NET assembly into memory directly from bytes stored in a Windows register.

This feature enables to reveal hidden artifacts encrypted within PE resources. Malicious artifacts are often encrypted to evade detection and obscure the true intent of the sample. Uncovering these artifacts is essential, as they typically contain critical data (as C2 information) or payloads. By extracting them, the sandbox can deliver a deeper scan, with higher chance of identifying the most valuable IOCs.

This sample stores that encrypted artifacts using the XOR algorithm, simple but efficient to evade detection. By analyzing patterns in the encrypted data, the encryption key can be guessed, allowing to decrypt the hidden.