Archive

Archive for the ‘Antimalware Scan Interface (AMSI)’ Category

Stopping Active Directory attacks and other post-exploitation behavior with AMSI and machine learning

August 27th, 2020 No comments

When attackers successfully breach a target network, their typical next step is to perform reconnaissance of the network, elevate their privileges, and move laterally to reach specific machines or spread as widely as possible. For these activities, attackers often probe the affected network’s Active Directory, which manages domain authentication and permissions for resources. Attackers take advantage of users’ ability to enumerate and interact with the Active Directory for reconnaissance, which allows lateral movement and privilege escalation. This is a common attack stage in human-operated ransomware campaigns like Ryuk.

These post-exploitation activities largely rely on scripting engines like PowerShell and WMI because scripts provide attackers flexibility and enable them to blend into the normal hum of enterprise endpoint activity. Scripts are lightweight, can be disguised and obfuscated relatively easily, and can be run fileless by loading them directly in memory through command-line or interacting with scripting engines in memory.

Antimalware Scan Interface (AMSI) helps security software to detect such malicious scripts by exposing script content and behavior. AMSI integrates with scripting engines on Windows 10 as well as Office 365 VBA to provide insights into the execution of PowerShell, WMI, VBScript, JavaScript, and Office VBA macros. Behavioral blocking and containment capabilities in Microsoft Defender Advanced Threat Protection (ATP) take full advantage of AMSI’s visibility into scripts and harness the power of machine learning and cloud-delivered protection to detect and stop malicious behavior. In the broader delivery of coordinated defense, the AMSI-driven detection of malicious scripts on endpoints helps Microsoft Threat Protection, which combines signals from Microsoft Defender ATP and other solutions in the Microsoft 365 security portfolio, to detect cross-domain attack chains.

On endpoints, performance-optimized machine learning models inspect script content and behavior through AMSI. When scripts run and malicious or suspicious behavior is detected, features are extracted from the content, including expert features, features selected by machine learning, and fuzzy hashes. The lightweight client machine learning models make inferences on the content. If the content is classified as suspicious, the feature description is sent to the cloud for full real-time classification. In the cloud, heavier counterpart machine learning models analyze the metadata and uses additional signals like file age, prevalence, and other such information to determine whether the script should be blocked or not.

These pairs of AMSI-powered machine learning classifiers, one pair for each scripting engine, allow Microsoft Defender ATP to detect malicious behavior and stop post-exploitation techniques and other script-based attacks, even after they have started running. In this blog, we’ll discuss examples of Active Directory attacks, including fileless threats, foiled by AMSI machine learning.

Diagram showing pairs of machine learning models on the endpoint and in the cloud using AMSI to detect malicious scripts

Figure 1. Pair of AMSI machine learning models on the client and in the cloud

Blocking BloodHound attacks

BloodHound is a popular open-source tool for enumerating and visualizing the domain Active Directory and is used by red teams and attackers as a post-exploitation tool. The enumeration allows a graph of domain devices, users actively signed into devices, and resources along with all their permissions. Attackers can discover and abuse weak permission configurations for privilege escalation by taking over other user accounts or adding themselves to groups with high privileges, or for planning their lateral movement path to their target privileges. Attackers, including those behind human-operated ransomware campaigns such as Ryuk, use BloodHound as part of their attacks.

To work, BloodHound uses a component called SharpHound to enumerate the domain and collect various categories of data: local admin collection, group membership collection, session collection, object property collection, ACL collection, and trust collection. This enumeration would typically then be exfiltrated to be visualized and analysed by the attacker as part of planning their next steps. SharpHound performs the domain enumeration and is officially published as a fileless PowerShell in-memory version, as well as a file-based executable tool version. It is critical to identify the PowerShell fileless variant enumeration if it is active on a network.

Code snippet of the SharpHound ingestor

Figure 2. SharpHound ingestor code snippets

When the SharpHound fileless PowerShell ingestor is run in memory, whether by a pen tester or an attacker, AMSI sees its execution buffer. The machine learning model on the client featurizes this buffer and sends it to the cloud for final classification.

Code snippet of SharpHound ingestor showing featurized details

Figure 3. Sample featurized SharpHound ingestor code

The counterpart machine learning model in the cloud analyzes the metadata, integrates other signals, and returns a verdict. Malicious scripts are detected and stopped on endpoints in real time:

Screenshot of Microsoft Defender Antivirus alert for detection of SharpHound

Figure 4. Microsoft Defender Antivirus detection of SharpHound

Detections are reported in Microsoft Defender Security Center, where SOC analysts can use Microsoft Defender ATP’s rich set of tools to investigate and respond to attacks:

Screenshot of Microsoft Defender Security Center showing detection of SharpHound

Figure 5. Microsoft Defender Security Center alert showing detection of SharpHound

This protection is provided by AI that has learned to identify and block these attacks automatically, and that will continue to adapt and learn new attack methods we observe.

Stopping Kerberoasting

Kerberoasting, like BloodHound attacks, is a technique for stealing credentials used by both red teams and attackers. Kerberoasting attacks abuse the Kerberos Ticket Granting Service (TGS) to gain access to accounts, typically targeting domain accounts for lateral movement.

Kerberoasting attacks involve scanning an Active Directory environment to generate a list of user accounts that have Kerberos Service Principal Name (SPN). Attackers then request these SPN to grant Kerberos Service Tickets to these accounts. The tickets are dumped from memory using various tools like Mimikatz and then exfiltrated for offline brute forcing on the encrypted segment of the tickets. If successful, attackers can identify the passwords associated with the accounts, which they then use to remotely sign into machines or access resources.

All the Kerberoasing attack steps leading to the hash extraction can be accomplished using a single PowerShell (Invoke-Kerberoast.ps1), and has been integrated into popular post-exploitation frameworks like PowerSploit and PowerShell Empire:

Figure 6. Single command line to download and execute Kerberoasting to extract user password hashes

Code snippet of Kerberoasting

Figure 7. Kerberoasting code

Because AMSI has visibility into PowerShell scripts, when the Invoke-Kerberoast.ps1 is run, AMSI allows for inspection of the PowerShell content during runtime. This buffer is featurized and analyzed by client-side machine learning models, and sent to the cloud for real-time ML classification.

Code snippet of Kerberoasting showing featurized details

Figure 8. Sample featurized Kerberoasting code

Microsoft Defender ATP raises an alert for the detection of Invoke-Kerberoast.ps1:

Figure 9. Microsoft Defender Security Center alert showing detection of Invoke-Kerberoast.ps1

Training the machine learning models

To ensure continued high-quality detection of threats, the AMSI machine learning models are trained per scripting engine using real-time protection data and threat investigations.

Featurization is key to machine learning models making intelligent decisions about whether content is malicious or benign. For behavior-based script logs, we extract the set of libraries, COM object, and function names used by the script. Learning the most important features within the script content is performed through a combination of character ngramming the script or behavior log, followed by semi-asynchronous stochastic dual coordinate ascent (SA-SDCA) algorithm with L1 regularization feature trimming to learn and deploy the most important character ngram features.

On top of the same features used to train the client models, other complex features used to train the cloud modes include fuzzy hashes, cluster hashes, partial hashes, and more. In addition, the cloud models have access to other information like age, prevalence, global file information, reputation and others, which allow cloud models to make more accurate decisions for blocking.

Conclusion: Broad visibility informs AI-driven protections

Across Microsoft, AI and machine learning protection technologies use Microsoft’s broad visibility into various surfaces to identify new and unknown threats. Microsoft Threat Protection uses these machine learning-driven protections to detect threats across endpoints, email and data, identities, and apps.

On endpoints, Microsoft Defender ATP uses multiple next-generation protection engines that detect a wide range of threats. One of these engines uses insights from AMSI and pairs of machine learning models on the client and in the cloud working together to detect and stop malicious scripts post-execution.

These pairs of AMSI models, one pair for each scripting engine, are part of the behavior-based blocking and containment capabilities in Microsoft Defender ATP, which are designed to detect and stop threats even after they have started running. When running, threats are exposed and can’t hide behind encryption or obfuscation. This adds another layer of protection for instances where sophisticated threats are able to slip through pre-execution defenses.

Diagram showing different next-generation protection engines on the client and in the cloud

Figure 10. Microsoft Defender ATP next-generation protection engines

In this blog post, we showed how these AMSI-driven behavior-based machine learning protections are critical in detecting and stopping post-exploitation activities like BloodHound-based and Kerberoasting attacks, which employ evasive malicious scripts, including fileless components. With AMSI, script content and behavior are exposed, allowing Microsoft Defender ATP to foil reconnaissance activities and prevent attacks from progressing.

To learn more about behavior-based blocking and containment, read the following blog posts:

 

Ankit Garg and Geoff McDonald

Microsoft Defender ATP Research Team

The post Stopping Active Directory attacks and other post-exploitation behavior with AMSI and machine learning appeared first on Microsoft Security.

Out of sight but not invisible: Defeating fileless malware with behavior monitoring, AMSI, and next-gen AV

September 27th, 2018 No comments

Consider this scenario: Two never-before-seen, heavily obfuscated scripts manage to slip past file-based detection and dynamically load an info-stealing payload into memory. The scripts are part of a social engineering campaign that tricks potential victims into running the scripts, which use the file names install_flash_player.js and BME040429CB0_1446_FAC_20130812.XML.PDF.js, to distribute and run the payload.

The payload is sophisticated and particularly elusive, given that it:

  • Doesnt touch the disk, and does not trigger antivirus file scanning
  • Is loaded in the context of the legitimate process that executed the scripts (i.e., wscript.exe)
  • Leaves no traces on the disk, such that forensic analysis finds limited evidence

These are markers of a fileless threat. Still, Windows Defender Advanced Threat Protection (Windows Defender ATP) antivirus capabilities detect the payload, stopping the attack in its tracks. How is this possible?

In this scenario, Antimalware Scan Interface (AMSI) facilitates detection. AMSI is an open interface that allows antivirus solutions to inspect script behavior by exposing script contents in a form that is both unencrypted and unobfuscated.

AMSI is part of the range of dynamic next-gen features that enable antivirus capabilities in Windows Defender ATP to go beyond file scanning. These features, which also include behavior monitoring, memory scanning, and boot sector protection, catch a wide spectrum of threats, including new and unknown (like the two scripts described above), fileless threats (like the payload), and other sophisticated malware.

Generically detecting fileless techniques

The two aforementioned obfuscated scripts are actual malware detected and blocked in the wild by antivirus capabilities in Windows Defender ATP. Removing the first layer of obfuscation reveals a code that, while still partially obfuscated, showed some functions related to a fileless malware technique called Sharpshooter. We found the two scripts, which were variants of the same malware, not long after the Sharpshooter technique was documented and published by MDSec in 2017.

The Sharpshooter technique allows an attacker to use a script to execute a .NET binary directly from memory without ever needing to reside on the disk. This technique provides a framework that can enable attackers to easily repackage the same binary payload within a script. As demonstrated by the example of the two scripts, files that use the Sharpshooter technique can then be used in social engineering attacks to lure users into running the script to deliver a fileless payload.

Screenshot of obfuscated scriptFigure 1. Obfuscated code from install_flash_player.js script

Screenshot of the script which contains functions typically used in the Sharpshooter technique

Figure 2. After de-obfuscation, the script contains functions typically used in the Sharpshooter technique

When the Sharpshooter technique became public, we knew it was only a matter time before it would be used it in attacks. To protect customers from such attacks, we implemented a detection algorithm based on runtime activity rather than on the static script. In other words, the detection is effective against the Sharpshooter technique itself, thus against new and unknown threats that implement the technique. This is how Windows Defender ATP blocked the two malicious scripts at first sight, preventing the fileless payload from being loaded.

The detection algorithm leverages AMSI support in scripting engines and targets a generic malicious behavior (a fingerprint of the malicious fileless technique). Script engines have the capability to log the APIs called by a script at runtime. This API logging is dynamic and is therefore not hindered by obfuscation: a script can hide its code, but it cannot hide its behavior. The log can then be scanned by antivirus solutions via AMSI when certain dangerous APIs (i.e., triggers) are invoked.

This is the dynamic log generated by the scripts and detected by Windows Defender ATP at runtime via AMSI:

Screenshot of the dynamic AMSI log generated during the execution of the Sharpshooter techniqueFigure 3. Dynamic AMSI log generated during the execution of the Sharpshooter technique in the two malicious scripts

Using this AMSI-aided detection, Windows Defender ATP disrupted two distinct malware campaigns in June, as well as the steady hum of daily activities.

Windows Defender ATP telemetry shows two Sharpshooter campaigns in JuneFigure 4. Windows Defender ATP telemetry shows two Sharpshooter campaigns in June

Furthermore, generically detecting the Sharpshooter technique allowed us to discover a particularly sophisticated and interesting attack. Windows Defender ATPs endpoint and detection response capabilities caught a VBScript file that used the Sharpshooter technique.

Sample Windows Defender ATP alert showing how detection of the Sharpshooter technique by Windows Defender AV is surfaced in Windows Defender Security CenterFigure 5. Sample Windows Defender ATP alert showing how detection of the Sharpshooter technique by Windows Defender AV is surfaced in Windows Defender Security Center

We analyzed the script and extracted the fileless payload, a very stealthy .NET executable. The malware payload downloads data from its command-and-control (C&C) server via the TXT records of DNS queries. In particular, it downloads the initialization vector and decryption key necessary to decode the core of the malware. The said core is also fileless because its executed directly in memory without being written on the disk. Thus, this attack leveraged two fileless stages.

Screenshot showing that the core component of the malware is decrypted and executed from memoryFigure 6. The core component of the malware is decrypted and executed from memory

Our investigation into the incident turned up enough indicators for us to conclude that this was likely a penetration testing exercise or a test involving running actual malware, and not a real targeted attack.

Nonetheless, the use of fileless techniques and the covert network communication hidden in DNS queries make this malware similar in nature to sophisticated, real-world attacks. It also proved the effectiveness of the dynamic protection capabilities of Windows Defender ATP. In a previous blog post, we documented how such capabilities allow Windows Defender ATP to catch KRYPTON attacks and other high-profile malware.

Upward trend in fileless attacks and living off the land

Removing the need for files is the next progression of attacker techniques. Antivirus solutions have become very efficient in detecting malicious executables. Real-time protection gives visibility on each new file that lands on the disk. Furthermore, file activity leaves a trail of evidence that can be retrieved during forensic analysis. That’s why we are seeing an increase in attacks that use of malware with fileless techniques.

At a high level, a fileless malware runs its main payload directly in memory without having to drop the executable file on the disk first. This differs from traditional malware, where the payload always requires some initial executable or DLL to carry out its tasks. A common example is the Kovter malware, which stores its executable payload entirely in registry keys. Going fileless allows the attackers to avoid having to rely on physical files and improve stealth and persistence.

For attackers, building fileless attacks poses some challenges; in primis: how do you execute code if you don’t have a file? Attackers found an answer in the way they infect other components to achieve execution within these components environment. Such components are usually standard, legitimate tools that are present by default on a machine and whose functionality can be abused to accomplish malicious operations.

This technique is usually referred to as “living off the land”, as malware only uses resources already available in the operating system. An example is the Trojan:Win32/Holiks.A malware abusing the mshta.exe tool:

Trojan:Win32/Holiks.A is abusing mshta.exe to execute a script from command-lineFigure 7. Trojan:Win32/Holiks.A is abusing mshta.exe to execute a script from command-line

The malicious script resides only in the command line; it loads and executes further code from a registry key. The whole execution happens within the context of the mshta.exe process, which is a clean executable and tends to be trusted as a legitimate component of the operating system. Other similar tools, such as cmstp.exe, regsvr32.exe, powershell.exe, odbcconf.exe, rundll3.exe, just to name a few, have been abused by attackers. Of course, the execution is not limited to scripts; the tools may allow the execution of DLLs and executables, even from remote locations in some cases.

By living off the land, fileless malware can cover its tracks: no files are available to the antivirus for scanning and only legitimate processes are executed. Windows Defender ATP overcomes this challenge by monitoring the behavior of the system for anomalies or known patterns of malicious usage of legitimate tools. For example, Trojan:Win32/Powemet.A!attk is a generic behavior-based detection designed to prevent attacks that leverage the regsvr32.exe tool to run malicious scripts.

Antivirus capabilities Windows Defender ATP blocking legitimate regsvr32 tool abused to download and run a malicious remote scriptFigure 8. Antivirus capabilities in Windows Defender ATP blocking legitimate regsvr32 tool abused to download and run a malicious remote script

What exactly is fileless?

The term fileless suggests that a threat that does not come in a file, such as a backdoor that lives only in the memory of a machine. However, theres no generally accepted definition. The term is used broadly; its also used to describe malware families that do rely on files in order to operate. In the Sharpshooter example, while the payload itself is fileless, the entry point relies on scripts that need to be dropped on the targets machine and executed. This, too, is considered a fileless attack.

Given that attacks involve several stages for functionalities like execution, persistence, information theft, lateral movement, communication with command-and-control, etc., some parts of the attack chain may be fileless, while others may involve the filesystem in some form or another.

To shed light on this loaded term, we grouped fileless threats into different categories.

Taxonomy of fileless threats

Figure 9. Taxonomy of fileless threats

We can classify fileless threats by their entry point (i.e., execution/injection, exploit, hardware), then the form of entry point (e.g., file, script, etc.), and finally by the host of the infection (e.g., Flash, Java, documents).

From this classification, we can glean three big types of fileless threats based on how much fingerprint they may leave on infected machines.

  • Type I: No file activity performed. A completely fileless malware can be considered one that never requires writing a file on the disk.
  • Type II: No files written on disk, but some files are used indirectly. There are other ways that malware can achieve fileless presence on a machine without requiring significant engineering effort. Fileless malware of this type do not directly write files on the file system, but they can end up using files indirectly.
  • Type III: Files required to achieve fileless persistence. Some malware can have some sort of fileless persistence but not without using files in order to operate.

Having described the broad categories, we can now dig into the details and provide a breakdown of the infection hosts. This comprehensive classification covers the panorama of what is usually referred to as fileless malware. It drives our efforts to research and develop new protection features that neutralize classes of attacks and ensure malware does not get the upper hand in the arms race.

Exploits Hardware Execution or injection

  • File-based (Type III: executable, Flash, Java, documents)
  • Network-based (Type I)

  • Device-based (Type I: network card, hard disk)
  • CPU-based (Type I)
  • USB-based (Type I)
  • BIOS-based (Type I)
  • Hypervisor-based (Type I)

  • File-based (Type III: executables, DLLs, LNK files, scheduled tasks)
  • Macro-based (Type III: Office documents)
  • Script-based (Type II: file, service, registry, WMI repo, shell)
  • Disk-based (Type II: Boot Record)

For a detailed description and examples of these categories, visit this comprehensive page on fileless threats.

Defeating fileless malware with next-gen protection

File-based inspection is ineffective against fileless malware. Antivirus capabilities in Windows Defender ATP use defensive layers based on dynamic behavior and integrate with other Windows technologies to detect and terminate threat activity at runtime.

Windows Defender ATPs next-gen dynamic defenses have become of paramount importance in protecting customers from the increasingly sophisticated attacks that fileless malware exemplifies. In a previous blog post we described some of the offensive and defensive technologies related to fileless attacks and how these solutions help protect our customers. Evolving from the file-centric scanning model, Windows Defender ATP uses a generic and more powerful behavior-centric detection model to neutralize generic malicious behaviors and thus take out entire classes of attack.

AMSI

Antimalware Scan Interface (AMSI) is an open framework that applications can use to request antivirus scans of any data. Windows leverages AMSI extensively in JavaScript, VBScript, and PowerShell. In addition, Office 365 client applications integrates with AMSI, enabling antivirus and other security solutions to scan macros and other scripts at runtime to check for malicious behavior. In the example above, we have shown how AMSI can be a powerful weapon to fight fileless malware.

Windows Defender ATP has implemented AMSI provider and consumes all AMSI signals for protection, these signals are especially effective against obfuscation. It has led to the disruption of malware campaigns like Nemucod. During a recent investigation, we stumbled upon some malicious scripts that were heavily obfuscated. We collected three samples that were evading static signatures and are a mixture of barely recognizable script code and binary junk data.

Heavy obfuscation of three different samples of TrojanDownloader:Script/Nemucod.JACFigure 10. Heavy obfuscation of three different samples of TrojanDownloader:Script/Nemucod.JAC.

However, after manual de-obfuscation, it turned out that these samples decode and execute the same .js script payload, a known downloader:

A portion of the second stage downloader decrypted by Nemucod.JACFigure 11: A portion of the second stage downloader decrypted by Nemucod.JAC

The payload does not have any obfuscation and is very easy to detect, but it never touches the disk and so could evade file-based detection. However, the scripting engine is capable of intercepting the attempt to execute the decoded payload and ensuring that the payload is passed to the installed antivirus via AMSI for inspection. Windows Defender ATP has visibility on the real payload as its decoded at runtime and can easily recognize known patterns and block the attack before it deals any damage.

Instead of writing a generic detection algorithm based on the obfuscation patterns in the samples, we trained an ML model on this behavior log and wrote heuristic detection to catch the decrypted scripts inspected via AMSI. The results proved effective, catching new and unknown variants, protecting almost two thousand machines in a span of two months. Traditional detection would not have been as effective.

Nemucod.JAC attack campaigns caught via AMSIFigure 12. Nemucod.JAC attack campaigns caught via AMSI

Behavior monitoring

Windows Defender ATPs behavior monitoring engine provides an additional layer of antivirus protection against fileless malware. The behavior monitoring engine filters suspicious API calls. Detection algorithms can then match dynamic behaviors that use particular sequences of APIs with specific parameters and block processes that expose known malicious behaviors. Behavior monitoring is useful not only for fileless malware, but also for traditional malware where the same malicious code base gets continuously repacked, encrypted, or obfuscated. Behavior monitoring proved effective against WannaCry, which was distributed through the DoublePulsar backdoor and can be categorized as a very dangerous Type I fileless malware. While several variants of the WannaCry binaries were released in attack waves, the behavior of the ransomware remained the same, allowing antivirus capabilities in Windows Defender ATP to block new versions of the ransomware.

Behavior monitoring is particularly useful against fileless attacks that live off the land. The PowerShell reverse TCP payload from Meterpreter is an example: it can be run completely on a command line and can provide a PowerShell session to a remote attacker.

Example of a possible command line generated by MeterpreterFigure 13. Example of a possible command line generated by Meterpreter

Theres no file to scan in this attack, but through behavior monitoring in its antivirus capabilities, Windows Defender ATP can detect the creation of the PowerShell process with the particular command line required. Behavior monitoring detects and blocks numerous attacks like this on a daily basis.

Detections of the PowerShell reverse TCP payloadFigure 14. Detections of the PowerShell reverse TCP payload

Beyond looking at events by process, behavior monitoring in Windows Defender ATP can also aggregate events across multiple processes, even if they are sparsely connected via techniques like code injection from one process to another (i.e., not just parent-child processes). Moreover, it can persist and orchestrate sharing of security signals across Windows Defender ATP components (e.g., endpoint detection and response) and trigger protection through other parts of the layered defenses.

Behavior monitoring across multiple processes is not only an effective protection against fileless malware; its also a tool to catch attack techniques in generic ways. Here is another example where multi process behavior monitoring in action, Pyordono.A is a detection based on multi-process events and is aimed at blocking scripting engines (JavaScript, VBScript, Office macros) that try to execute cmd.exe or powershell.exe with suspicious parameters. Windows Defender ATP telemetry shows this detection algorithm protecting users from several campaigns.

Pyordono.A technique detected in the wildFigure 15. Pyordono.A technique detected in the wild

Recently, we saw a sudden increase in Pyordono.A encounters, reaching levels way above the average. We investigated this anomaly and uncovered a widespread campaign that used malicious Excel documents and targeted users in Italy from September 8 to 12.

Screenshot of malicious Excel document with instructions in Italian to click Enable contentFigure 16. Malicious Excel document with instructions in Italian to click Enable content

The document contains a malicious macro and uses social engineering to lure potential victims into running the malicious code. (Note: We have recently integrated Office 365 clients apps with AMSI, enabling antivirus solutions to scan macros at runtime to check for malicious content).

The obfuscated macro code attempts to run an obfuscated Cmd command which in turns executes an obfuscated Powershell script. In the end, the Ursnif trojan is delivered.Figure 17. The obfuscated macro code attempts to run an obfuscated Cmd command which in turns executes an obfuscated Powershell script. In the end, the Ursnif trojan is delivered.

The macro makes use of obfuscation to execute a cmd command, which is also obfuscated. The cmd command executes a PowerShell script that in turn downloads additional data and delivers the payload, infostealing Ursnif. We recently reported a small-scale Ursnif campaign that targeted small businesses in specific US cities. Through multi-process behavior monitoring, Windows Defender ATP detected and blocked the new campaign targeting users in Italy using a generic detection algorithm without prior knowledge of the malware.

Memory scanning

Antivirus capabilities in Windows Defender ATP also employ memory scanning to detect the presence of malicious code in the memory of a running process. Even if malware can run without the use of a physical file, it does need to reside in memory in order to operate and is therefore detectable by means of memory scanning. An example is the GandCrab ransomware, which was reported to have become fileless. The payload DLL is encoded in a string, then decoded and run dynamically via PowerShell. The DLL itself is never dropped on the disk. Using memory scanning, Windows Defender ATP can scan the memory of running processes and detect known patterns of the ransomware run from the stealthy DLL.

Memory scanning, in conjunction with behavior monitoring and other dynamic defenses, helped Windows Defender ATP to disrupt a massive Dofoil campaign. Dofoil, a known nasty downloader, uses some sophisticated techniques to evade detection, including process hollowing, which allows the malware to execute in the context of a legitimate process (e.g., explorer.exe). To this day, memory scanning detects Dofoil activities.

Detections of the memory-resident Dofoil payloadFigure 18. Detections of the memory-resident Dofoil payload

Memory scanning is a versatile tool: when suspicious APIs or behavior monitoring events are observed at runtime, antivirus capabilities in Windows Defender ATP trigger a memory scan in key points it is more likely to observe (and detect) a payload that has been decoded and may be about to run. This gives Windows Defender ATP granular control on which actions are more interesting and may require more attention. Every day, memory scanning allows Windows Defender ATP to protect thousands of machines against active high-profile threats like Mimikatz and WannaCry.

Boot Sector protection

With Controlled folder access on Windows 10, Windows Defender ATP does not allow write operations to the boot sector, thus closing a dangerous fileless attack vector used by Petya, BadRabbit, and bootkits in general. Boot infection techniques can be suitable for fileless threats because it can allow malware to reside outside of the file system and gain control of the machine before the operating system is loaded. The use of rootkit techniques, like in the defunct Alureon malware (also known as TDSS or TDL-4), can then render the malware invisible and extremely difficult to detect and remove. With Controlled folder access, which is part of Windows Defender ATPs attack surface reduction capabilities, this entire class of infection technique has become a thing of the past.

Control Folder Access preventing a boot sector infection attempted by PetyaFigure 19. Control Folder Access preventing a boot sector infection attempted by Petya

Windows 10 in S mode: Naturally resistant to fileless attacks

Windows 10 in S mode comes with a preconfigured set of restrictions and policies that make it naturally protected against a vast majority of the fileless techniques (and against malware in general). Among the available security features, the following ones are particularly effective against fileless threats:

For executables: Only Microsoft-verified applications from the Microsoft Store are allowed to run. Furthermore, Device Guard provides User Mode Code Integrity (UMCI) to prevent the loading of unsigned binaries.

For scripts: Scripting engines are not allowed to run (including JavaScript, VBScript, and PowerShell).

For macros: Office 365 does not allow the execution of macros in documents from the internet (for example, documents that are downloaded or received as attachment in emails from outside the organization).

For exploits: Exploit protection and Attack surface reduction rules are also available on Windows 10 in S mode as a consistent barrier against exploitation.

With these restrictions in place, Windows 10 in S mode devices are in a robust, locked down state, removing crucial attack vectors used by fileless malware.

Conclusion

As antivirus solutions become better and better at pinpointing malicious files, the natural evolution of malware is to shift to attack chains that use as few files as possible. While fileless techniques used to be employed almost exclusively in sophisticated cyberattacks, they are now becoming widespread in common malware, too.

At Microsoft, we actively monitor the security landscape to identify new threat trends and develop solutions that continuously enhance Windows security and mitigate classes of threats. We instrument durable generic detections that are effective against a wide range of threats. Through AMSI, behavior monitoring, memory scanning, and boot sector protection, we can inspect threats even with heavy obfuscation. Machine learning technologies in the cloud allow us to scale these protections against new and emerging threats.

Security solutions on Windows 10 integrate into a unified endpoint security platform in Windows Defender Advanced Threat Protection. Windows Defender ATP includes attack surface reduction, next-generation protection, endpoint protection and response, auto investigation and remediation, security posture, and advanced hunting capabilities. To test how Windows Defender ATP can help your organization detect, investigate, and respond to advanced attacks, sign up for a free trial.

Protections against fileless and other threats are shared across Microsoft 365, which integrate technologies in Windows, Office 365, and Azure. Through the Microsoft Intelligent Security Graph, security signals are shared and remediation is orchestrated across Microsoft 365.

 

 

Andrea Lelli
Windows Defender Research

 

 

 

 

 

 


Talk to us

Questions, concerns, or insights on this story? Join discussions at the Microsoft community and Windows Defender Security Intelligence.

Follow us on Twitter @WDSecurity and Facebook Windows Defender Security Intelligence.

The post Out of sight but not invisible: Defeating fileless malware with behavior monitoring, AMSI, and next-gen AV appeared first on Microsoft Secure.

Windows Defender ATP machine learning and AMSI: Unearthing script-based attacks that ‘live off the land’

December 4th, 2017 No comments

Data center

Scripts are becoming the weapon of choice of sophisticated activity groups responsible for targeted attacks as well as malware authors who indiscriminately deploy commodity threats.

Scripting engines such as JavaScript, VBScript, and PowerShell offer tremendous benefits to attackers. They run through legitimate processes and are perfect tools for living off the landstaying away from the disk and using common tools to run code directly in memory. Often part of the operating system, scripting engines can evaluate and execute content from the internet on-the-fly. Furthermore, integration with popular apps make them effective vehicles for delivering malicious implants through social engineering as evidenced by the increasing use of scripts in spam campaigns.

Malicious scripts are not only used as delivery mechanisms. We see them in various stages of the kill chain, including during lateral movement and while establishing persistence. During these latter stages, the scripting engine of choice is clearly PowerShellthe de facto scripting standard for administrative tasks on Windowswith the ability to invoke system APIs and access a variety of system classes and objects.

While the availability of powerful scripting engines makes scripts convenient tools, the dynamic nature of scripts allows attackers to easily evade analysis and detection by antimalware and similar endpoint protection products. Scripts are easily obfuscated and can be loaded on-demand from a remote site or a key in the registry, posing detection challenges that are far from trivial.

Windows 10 provides optics into script behavior through Antimalware Scan Interface (AMSI), a generic, open interface that enables Windows Defender Antivirus to look at script contents the same way script interpreters doin a form that is both unencrypted and unobfuscated. In Windows 10 Fall Creators Update, with knowledge from years analyzing script-based malware, weve added deep behavioral instrumentation to the Windows script interpreter itself, enabling it to capture system interactions originating from scripts. AMSI makes this detailed interaction information available to registered AMSI providers, such as Windows Defender Antivirus, enabling these providers to perform further inspection and vetting of runtime script execution content.

This unparalleled visibility into script behavior is capitalized further through other Windows 10 Fall Creators Update enhancements in both Windows Defender Antivirus and Windows Defender Advanced Threat Protection (Windows Defender ATP). Both solutions make use of powerful machine learning algorithms that process the improved optics, with Windows Defender Antivirus delivering enhanced blocking of malicious scripts pre-breach and Windows Defender ATP providing effective behavior-based alerting for malicious post-breach script activity.

In this blog, we explore how Windows Defender ATP, in particular, makes use of AMSI inspection data to surface complex and evasive script-based attacks. We look at advanced attacks perpetrated by the highly skilled KRYPTON activity group and explore how commodity malware like Kovter abuses PowerShell to leave little to no trace of malicious activity on disk. From there, we look at how Windows Defender ATP machine learning systems make use of enhanced insight about script characteristics and behaviors to deliver vastly improved detection capabilities.

KRYPTON: Highlighting the resilience of script-based attacks

Traditional approaches for detecting potential breaches are quite file-centric. Incident responders often triage autostart entries, sorting out suspicious files by prevalence or unusual name-folder combinations. With modern attacks moving closer towards being completely fileless, it is crucial to have additional sensors at relevant choke points.

Apart from not having files on disk, modern script-based attacks often store encrypted malicious content separately from the decryption key. In addition, the final key often undergoes multiple processes before it is used to decode the actual payload, making it is impossible to make a determination based on a single file without tracking the actual invocation of the script. Even a perfect script emulator would fail this task.

For example, the activity group KRYPTON has been observed hijacking or creating scheduled tasksthey often target system tasks found in exclusion lists of popular forensic tools like Autoruns for Windows. KRYPTON stores the unique decryption key within the parameters of the scheduled task, leaving the actual payload content encrypted.

To illustrate KRYPTON attacks, we look at a tainted Microsoft Word document identified by John Lambert and the Office 365 Advanced Threat Protection team.

KRYPTON lure document

Figure 1. KRYPTON lure document

To live off the land, KRYPTON doesnt drop or carry over any traditional malicious binaries that typically trigger antimalware alerts. Instead, the lure document contains macros and uses the Windows Scripting Host (wscript.exe) to execute a JavaScript payload. This script payload executes only with the right RC4 decryption key, which is, as expected, stored as an argument in a scheduled task. Because it can only be triggered with the correct key introduced in the right order, the script payload is resilient against automated sandbox detonations and even manual inspection.

KRYPTON script execution chain through wscript.exe

Figure 2. KRYPTON script execution chain through wscript.exe

Exposing actual script behavior with AMSI

AMSI overcomes KRYPTONs evasion mechanisms by capturing JavaScript API calls after they have been decrypted and ready to be executed by the script interpreter. The screenshot below shows part of the exposed content from the KRYPTON attack as captured by AMSI.

Part of the KRYPTON script payload captured by AMSI and sent to the cloud for analysis

Figure 3. Part of the KRYPTON script payload captured by AMSI and sent to the cloud for analysis

By checking the captured script behavior against indicators of attack (IoAs) built up by human experts as well as machine learning algorithms, Windows Defender ATP effortlessly flags the KRYPTON scripts as malicious. At the same time, Windows Defender ATP provides meaningful contextual information, including how the script is triggered by a malicious Word document.

Windows Defender ATP machine learning detection of KRYPTON script captured by AMSI

Figure 4. Windows Defender ATP machine learning detection of KRYPTON script captured by AMSI

PowerShell use by Kovter and other commodity malware

Not only advanced activity groups like KRYPTON are shifting from binary executables to evasive scripts. In the commodity space, Kovter malware uses several processes to eventually execute its malicious payload. This payload resides in a PowerShell script decoded by a JavaScript (executed by wscript.exe) and passed to powershell.exe as an environment variable.

Windows Defender ATP machine learning alert for the execution of the Kovter script-based payload

Figure 5. Windows Defender ATP machine learning alert for the execution of the Kovter script-based payload

By looking at the PowerShell payload content captured by AMSI, experienced analysts can easily spot similarities to PowerSploit, a publicly available set of penetration testing modules. While such attack techniques involve file-based components, they remain extremely hard to detect using traditional methods because malicious activities occur only in memory. Such behavior, however, is effortlessly detected by Windows Defender ATP using machine learning that combines detailed AMSI signals with signals generated by PowerShell activity in general.

Part of the Kovter script payload captured by AMSI and sent to the cloud for analysis

Figure 6. Part of the Kovter script payload captured by AMSI and sent to the cloud for analysis

Fresh machine learning insight with AMSI

While AMSI provides rich information from captured script content, the highly variant nature of malicious scripts continues to make them challenging targets for detection. To efficiently extract and identify new traits differentiating malicious scripts from benign ones, Windows Defender ATP employs advanced machine learning methods.

As outlined in our previous blog, we employ a supervised machine learning classifier to identify breach activity. We build training sets based on malicious behaviors observed in the wild and normal activities on typical machines, augmenting that with data from controlled detonations of malicious artifacts. The diagram below conceptually shows how we capture malicious behaviors in the form of process trees.

Process tree augmented by instrumentation for AMSI data

Figure 7. Process tree augmented by instrumentation for AMSI data

As shown in the process tree, the kill chain begins with a malicious document that causes Microsoft Word (winword.exe) to launch PowerShell (powershell.exe). In turn, PowerShell executes a heavily obfuscated script that drops and executes the malware fhjUQ72.tmp, which then obtains persistence by adding a run key to the registry. From the process tree, our machine learning systems can extract a variety of features to build expert classifiers for areas like registry modification and file creation, which are then converted into numeric scores that are used to decide whether to raise alerts.

With the instrumentation of AMSI signals added as part of the Windows 10 Fall Creators Update (version 1709), Windows Defender ATP machine learning algorithms can now make use of insight into the unobfuscated script content while continually referencing machine state changes associated with process activity. Weve also built a variety of script-based models that inspect the nature of executed scripts, such as the count of obfuscation layers, entropy, obfuscation features, ngrams, and specific API invocations, to name a few.

As AMSI peels off the obfuscation layers, Windows Defender ATP benefits from growing visibility and insight into API calls, variable names, and patterns in the general structure of malicious scripts. And while AMSI data helps improve human expert knowledge and their ability to train learning systems, our deep neural networks automatically learn features that are often hidden from human analysts.

Machine-learning detections of JavaScript and PowerShell scripts

Figure 8. Machine learning detections of JavaScript and PowerShell scripts

While these new script-based machine learning models augment our expert classifiers, we also correlate new results with other behavioral information. For example, Windows Defender ATP correlates the detection of suspicious script contents from AMSI with other proximate behaviors, such as network connections. This contextual information is provided to SecOps personnel, helping them respond to incidents efficiently.

Machine learning combines VBScript content from AMSI and tracked network activity

Figure 9. Machine learning combines VBScript content from AMSI and tracked network activity

Detection of AMSI bypass attempts

With AMSI providing powerful insight into malicious script activity, attacks are more likely to incorporate AMSI bypass mechanisms that we group into three categories:

  • Bypasses that are part of the script content and can be inspected and alerted on
  • Tampering with the AMSI sensor infrastructure, which might involve the replacement of system files or manipulation of the load order of relevant DLLs
  • Patching of AMSI instrumentation in memory

The Windows Defender ATP research team proactively develops anti-tampering mechanisms for all our sensors. We have devised heuristic alerts for possible manipulation of our optics, designing these alerts so that they are triggered in the cloud before the bypass can suppress them.

During actual attacks involving CVE-2017-8759, Windows Defender ATP not only detected malicious post-exploitation scripting activity but also detected attempts to bypass AMSI using code similar to one identified by Matt Graeber.

Windows Defender ATP alert based on AMSI bypass pattern

Figure 10. Windows Defender ATP alert based on AMSI bypass pattern

AMSI itself captured the following bypass code for analysis in the Windows Defender ATP cloud.

AMSI bypass code sent to the cloud for analysis

Figure 11. AMSI bypass code sent to the cloud for analysis

Conclusion: Windows Defender ATP machine learning and AMSI provide revolutionary defense against highly evasive script-based attacks

Provided as an open interface on Windows 10, Antimalware Scan Interface delivers powerful optics into malicious activity hidden in encrypted and obfuscated scripts that are oftentimes never written to disk. Such evasive use of scripts is becoming commonplace and is being employed by both highly skilled activity groups and authors of commodity malware.

AMSI captures malicious script behavior by looking at script content as it is interpreted, without having to check physical files or being hindered by obfuscation, encryption, or polymorphism. At the endpoint, AMSI benefits local scanners, providing the necessary optics so that even obfuscated and encrypted scripts can be inspected for malicious content. Windows Defender Antivirus, specifically, utilizes AMSI to dynamically inspect and block scripts responsible for dropping all kinds of malicious payloads, including ransomware and banking trojans.

With Windows 10 Fall Creators Update (1709), newly added script runtime instrumentation provides unparalleled visibility into script behaviors despite obfuscation. Windows Defender Antivirus uses this treasure trove of behavioral information about malicious scripts to deliver pre-breach protection at runtime. To deliver post-breach defense, Windows Defender ATP uses advanced machine learning systems to draw deeper insight from this data.

Apart from looking at specific activities and patterns of activities, new machine learning algorithms in Windows Defender ATP look at script obfuscation layers, API invocation patterns, and other features that can be used to efficiently identify malicious scripts heuristically. Windows Defender ATP also correlates script-based indicators with other proximate activities, so it can deliver even richer contextual information about suspected breaches.

To benefit from the new script runtime instrumentation and other powerful security enhancements like Windows Defender Exploit Guard, customers are encourage to install Windows 10 Fall Creators Update.

Read the The Total Economic Impact of Microsoft Windows Defender Advanced Threat Protection from Forrester to understand the significant cost savings and business benefits enabled by Windows Defender ATP. To directly experience how Windows Defender ATP can help your enterprise detect, investigate, and respond to advance attacks, sign up for a free trial.

 

Stefan Sellmer, Windows Defender ATP Research

with

Shay Kels, Windows Defender ATP Research

Karthik Selvaraj, Windows Defender Research

 

Additional readings

 


Talk to us

Questions, concerns, or insights on this story? Join discussions at the Microsoft community and Windows Defender Security Intelligence.

Follow us on Twitter @WDSecurity and Facebook Windows Defender Security Intelligence.