June 5, 2025

Unpacking ClickFix: Darktrace’s detection of a prolific social engineering tactic

ClickFix is a social engineering technique that exploits human error through fake prompts, leading users to unknowingly run malicious commands. Learn how Darktrace detects and responds to such threats!

No items found.

Inside the SOC

Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.

No items found.

Jun 2025

What is ClickFix and how does it work?

Amid heightened security awareness, threat actors continue to seek stealthy methods to infiltrate target networks, often finding the human end user to be the most vulnerable and easily exploited entry point.

ClickFix baiting is an exploitation of the end user, making use of social engineering techniques masquerading as error messages or routine verification processes, that can result in malicious code execution.

Since March 2024, the simplicity of this technique has drawn attention from a range of threat actors, from individual cybercriminals to Advanced Persistent Threat (APT) groups such as APT28 and MuddyWater, linked to Russia and Iran respectively, introducing security threats on a broader scale [1]. ClickFix campaigns have been observed affecting organizations in across multiple industries, including healthcare, hospitality, automotive and government [2][3].

Actors carrying out these targeted attacks typically utilize similar techniques, tools and procedures (TTPs) to gain initial access. These include spear phishing attacks, drive-by compromises, or exploiting trust in familiar online platforms, such as GitHub, to deliver malicious payloads [2][3]. Often, a hidden link within an email or malvertisements on compromised legitimate websites redirect the end user to a malicious URL [4]. These take the form of ‘Fix It’ or fake CAPTCHA prompts [4].

From there, users are misled into believing they are completing a human verification step, registering a device, or fixing a non-existent issue such as a webpage display error. As a result, they are guided through a three-step process that ultimately enables the execution of malicious PowerShell commands:

Open a Windows Run dialog box [press Windows Key + R]
Automatically or manually copy and paste a malicious PowerShell command into the terminal [press CTRL+V]
And run the prompt [press ‘Enter’] [2]

Once the malicious PowerShell command is executed, threat actors then establish command and control (C2) communication within the targeted environment before moving laterally through the network with the intent of obtaining and stealing sensitive data [4]. Malicious payloads associated with various malware families, such as XWorm, Lumma, and AsyncRAT, are often deployed [2][3].

Attack timeline of ClickFix cyber attack

Based on investigations conducted by Darktrace’s Threat Research team in early 2025, this blog highlights Darktrace’s capability to detect ClickFix baiting activity following initial access.

Darktrace’s coverage of a ClickFix attack chain

Darktrace identified multiple ClickFix attacks across customer environments in both Europe, the Middle East, and Africa (EMEA) and the United States. The following incident details a specific attack on a customer network that occurred on April 9, 2025.

Although the initial access phase of this specific attack occurred outside Darktrace’s visibility, other affected networks showed compromise beginning with phishing emails or fake CAPTCHA prompts that led users to execute malicious PowerShell commands.

Darktrace’s visibility into the compromise began when the threat actor initiated external communication with their C2 infrastructure, with Darktrace / NETWORK detecting the use of a new PowerShell user agent, indicating an attempt at remote code execution.

Figure 1: Darktrace / NETWORK's detection of a device making an HTTP connection with new PowerShell user agent, indicating PowerShell abuse for C2 communications.

Download of Malicious Files for Lateral Movement

A few minutes later, the compromised device was observed downloading a numerically named file. Numeric files like this are often intentionally nondescript and associated with malware. In this case, the file name adhered to a specific pattern, matching the regular expression: /174(\d){7}/. Further investigation into the file revealed that it contained additional malicious code designed to further exploit remote services and gather device information.

‍

Figure 2: Darktrace / NETWORK's detection of a numeric file, one minute after the new PowerShell User Agent alert.

‍

The file contained a script that sent system information to a specified IP address using an HTTP POST request, which also processed the response. This process was verified through packet capture (PCAP) analysis conducted by the Darktrace Threat Research team.

By analyzing the body content of the HTTP GET request, it was observed that the command converts the current time to Unix epoch time format (i.e., 9 April 2025 13:26:40 GMT), resulting in an additional numeric file observed in the URI: /1744205200.

Figure 3: PCAP highlighting the HTTP GET request that sends information to the specific IP, 193.36.38[.]237, which then generates another numeric file titled per the current time.

Across Darktrace’s investigations into other customers' affected by ClickFix campaigns, both internal information discovery events and further execution of malicious code were observed.

Data Exfiltration

By following the HTTP stream in the same PCAP, the Darktrace Threat Research Team assessed the activity as indicative of data exfiltration involving system and device information to the same command-and-control (C2) endpoint, , 193.36.38[.]237. This endpoint was flagged as malicious by multiple open-source intelligence (OSINT) vendors [5].

Further analysis of Darktrace’s Advanced Search logs showed that the attacker’s malicious code scanned for internal system information, which was then sent to a C2 server via an HTTP POST request, indicating data exfiltration

Figure 5: Advanced Search further highlights Darktrace's observation of the HTTP POST request, with the second numeric file representing data exfiltration.

Actions on objectives

Around ten minutes after the initial C2 communications, the compromised device was observed connecting to an additional rare endpoint, 188.34.195[.]44. Further analysis of this endpoint confirmed its association with ClickFix campaigns, with several OSINT vendors linking it to previously reported attacks [6].

In the final HTTP POST request made by the device, Darktrace detected a file at the URI /init1234 in the connection logs to the malicious endpoint 188.34.195[.]44, likely depicting the successful completion of the attack’s objective, automated data egress to a ClickFix C2 server.

Darktrace / NETWORK grouped together the observed indicators of compromise (IoCs) on the compromised device and triggered an Enhanced Monitoring model alert, a high-priority detection model designed to identify activity indicative of the early stages of an attack. These models are monitored and triaged 24/7 by Darktrace’s Security Operations Center (SOC) as part of the Managed Threat Detection service, ensuring customers are promptly notified of malicious activity as soon as it emerges.

Figure 6: Darktrace correlated the separate malicious connections that pertained to a single campaign.

Darktrace Autonomous Response

In the incident outlined above, Darktrace was not configured in Autonomous Response mode. As a result, while actions to block specific connections were suggested, they had to be manually implemented by the customer’s security team. Due to the speed of the attack, this need for manual intervention allowed the threat to escalate without interruption.

However, in a different example, Autonomous Response was fully enabled, allowing Darktrace to immediately block connections to the malicious endpoint (138.199.156[.]22) just one second after the initial connection in which a numerically named file was downloaded [7].

Figure 7: Darktrace Autonomous Response blocked connections to a suspicious endpoint following the observation of the numeric file download.

This customer was also subscribed to our Managed Detection and Response service, Darktrace’s SOC extended a ‘Quarantine Device’ action that had already been autonomously applied in order to buy their security team additional time for remediation.

Figure 8: Autonomous Response blocked connections to malicious endpoints, including 138.199.156[.]22, 185.250.151[.]155, and rkuagqnmnypetvf[.]top, and also quarantined the affected device. These actions were later manually reinforced by the Darktrace SOC.

Conclusion

ClickFix baiting is a widely used tactic in which threat actors exploit human error to bypass security defenses. By tricking end point users into performing seemingly harmless, everyday actions, attackers gain initial access to systems where they can access and exfiltrate sensitive data.

Darktrace’s anomaly-based approach to threat detection identifies early indicators of targeted attacks without relying on prior knowledge or IoCs. By continuously learning each device’s unique pattern of life, Darktrace detects subtle deviations that may signal a compromise. In this case, Darktrace's Autonomous Response, when operating in a fully autonomous mode, was able to swiftly contain the threat before it could progress further along the attack lifecycle.

Credit to Keanna Grelicha (Cyber Analyst) and Jennifer Beckett (Cyber Analyst)

Appendices

NETWORK Models

Device / New PowerShell User Agent
Anomalous Connection / New User Agent to IP Without Hostname
Anomalous Connection / Posting HTTP to IP Without Hostname
Anomalous Connection / Powershell to Rare External
Device / Suspicious Domain
Device / New User Agent and New IP
Anomalous File / New User Agent Followed By Numeric File Download (Enhanced Monitoring Model)
Device / Initial Attack Chain Activity (Enhanced Monitoring Model)

Autonomous Response Models

Antigena / Network::Significant Anomaly::Antigena Significant Anomaly from Client Block
Antigena / Network::Significant Anomaly::Antigena Enhanced Monitoring from Client Block
Antigena / Network::External Threat::Antigena File then New Outbound Block
Antigena / Network::External Threat::Antigena Suspicious File Block
Antigena / Network::Significant Anomaly::Antigena Alerts Over Time Block
Antigena / Network::External Threat::Antigena Suspicious File Block

IoC - Type - Description + Confidence

· 141.193.213[.]11 – IP address – Possible C2 Infrastructure

· 141.193.213[.]10 – IP address – Possible C2 Infrastructure

· 64.94.84[.]217 – IP address – Possible C2 Infrastructure

· 138.199.156[.]22 – IP address – C2 server

· 94.181.229[.]250 – IP address – Possible C2 Infrastructure

· 216.245.184[.]181 – IP address – Possible C2 Infrastructure

· 212.237.217[.]182 – IP address – Possible C2 Infrastructure

· 168.119.96[.]41 – IP address – Possible C2 Infrastructure

· 193.36.38[.]237 – IP address – C2 server

· 188.34.195[.]44 – IP address – C2 server

· 205.196.186[.]70 – IP address – Possible C2 Infrastructure

· rkuagqnmnypetvf[.]top – Hostname – C2 server

· shorturl[.]at/UB6E6 – Hostname – Possible C2 Infrastructure

· tlgrm-redirect[.]icu – Hostname – Possible C2 Infrastructure

· diagnostics.medgenome[.]com – Hostname – Compromised Website

· /1741714208 – URI – Possible malicious file

· /1741718928 – URI – Possible malicious file

· /1743871488 – URI – Possible malicious file

· /1741200416 – URI – Possible malicious file

· /1741356624 – URI – Possible malicious file

· /ttt – URI – Possible malicious file

· /1741965536 – URI – Possible malicious file

· /1.txt – URI – Possible malicious file

· /1744205184 – URI – Possible malicious file

· /1744139920 – URI – Possible malicious file

· /1744134352 – URI – Possible malicious file

· /1744125600 – URI – Possible malicious file

· /1[.]php?s=527 – URI – Possible malicious file

· 34ff2f72c191434ce5f20ebc1a7e823794ac69bba9df70721829d66e7196b044 – SHA-256 Hash – Possible malicious file

· 10a5eab3eef36e75bd3139fe3a3c760f54be33e3 – SHA-1 Hash – Possible malicious file

MITRE ATT&CK Mapping

Tactic – Technique – Sub-Technique

Spearphishing Link - INITIAL ACCESS - T1566.002 - T1566

Drive-by Compromise - INITIAL ACCESS - T1189

PowerShell - EXECUTION - T1059.001 - T1059

Exploitation of Remote Services - LATERAL MOVEMENT - T1210

Web Protocols - COMMAND AND CONTROL - T1071.001 - T1071

Automated Exfiltration - EXFILTRATION - T1020 - T1020.001

References

[1] https://www.logpoint.com/en/blog/emerging-threats/clickfix-another-deceptive-social-engineering-technique/

[2] https://www.proofpoint.com/us/blog/threat-insight/security-brief-clickfix-social-engineering-technique-floods-threat-landscape

[3] https://cyberresilience.com/threatonomics/understanding-the-clickfix-attack/

[4] https://www.group-ib.com/blog/clickfix-the-social-engineering-technique-hackers-use-to-manipulate-victims/

[5] https://www.virustotal.com/gui/ip-address/193.36.38.237/detection

[6] https://www.virustotal.com/gui/ip-address/188.34.195.44/community

[7] https://www.virustotal.com/gui/ip-address/138.199.156.22/detection

No items found.

Inside the SOC

No items found.

Specialist Security Researcher

Watch the NIS2 Webinar

Blog

Network

September 25, 2026

A Chain Reaction: Blockchain-Hosted Infostealer Campaign Targets Windows and macOS

Key Insights

Darktrace detected a blockchain-hosted infostealer campaign targeting Windows and macOS devices across multiple customer environments.
The campaign combined ClickFix social engineering with trusted services and decentralized blockchain infrastructure to support malware delivery and C2 activity.
Compromised devices were observed connecting to rare and unusual external endpoints, including DGA C2 domains, blockchain-related endpoints, and cryptocurrency mining infrastructure.
The activity was associated with information-stealing malware strains including Atomic macOS Stealer (AMOS), Lumma, Rhadamanthys, Vidar, and Phexia.
Darktrace identified anomalous device behavior, beaconing patterns, rare external connections, cryptomining activity, and suspicious TLS/SSL communications without relying solely on prior knowledge or static indicators of compromise.
The campaign highlights how attackers are increasingly using legitimate and decentralized infrastructure to make detection, disruption, and attribution more challenging for defenders.

The Infostealer Ecosystem

The information stealer malware ecosystem continues to grow in value for threat actors across the digital threat landscape. Infostealers are increasingly delivered through Malware-as-a-Service (MaaS) operating models, distributed through affiliate networks, and designed to withstand infrastructure takedowns. This resilience was demonstrated by the recent takedown of Lumma Stealer malicious domains by Microsoft’s Digital Crimes Unit (DCU) [1].

Infostealers are used to gather and exfiltrate sensitive information, including non-human identity (NHI) data, from compromised systems across cloud, Software-as-a-Service (SaaS), Virtual Private Network (VPN), and development environments. They can also support ransomware operations by expanding the credentials and access paths available to threat actors, contributing to the high volume of identity-based attacks observed across the broader threat landscape [2][3].

Darktrace’s Observations of ClickFix and Infostealers

Throughout 2026, Darktrace has observed multiple campaigns using ClickFix social engineering to trick users into carrying out malicious actions and downloading initial payloads, including information stealers. More recently, Darktrace’s Threat Research team identified a specific ClickFix campaign involving a blockchain-hosted infostealer targeting Windows and macOS devices.

Darktrace identified affected customer environments across Europe, the United States, Asia, and the Middle East where blockchain-hosted infostealer malware appears to have been delivered to compromised systems following likely ClickFix-driven initial access. Darktrace investigated the activity and found that decentralized blockchain infrastructure, alongside widely trusted legitimate services, was used to support malware delivery and information theft across Windows and macOS systems.

Following initial access, compromised systems established C2 communication, with C2 configuration and payloads hosted on public blockchain infrastructure. The ultimate objective appears to be credential and cryptocurrency theft through the deployment of information stealers such as Atomic macOS Stealer (AMOS), Lumma, Rhadamanthys, and Vidar [5][6][7].

Darktrace’s Investigation

Affected devices across the Darktrace customer base were observed making outbound connections to rare external endpoints in patterns consistent with beaconing and C2 activity. Darktrace primarily detected devices making repeated connections to algorithmically generated domains (DGA) such as hf98x4d[.]site [8]. In many cases, these domains were linked through open-source intelligence (OSINT) to information-stealing malware families including AMOS and Phexia [5][6][7][8][9].

In multiple cases, devices were also observed connecting to blockchain-related endpoints, such as polygon[.]drpc[.]org, as well as legitimate public services, including GitHub. The use of decentralized blockchain infrastructure and trusted services such as GitHub to facilitate malware distribution and C2 activity can make disruption and attribution significantly more difficult for defenders.

Darktrace alsodetected a significant proportion of impacted devices making outboundconnections to cryptocurrency mining infrastructure associated with thelegitimate open-source XMRig mining software and the HashVault mining pool,including pool.hashvault[.]pro and donate[.]ssl[.]xmrig[.]com, which wereabused by the attackers, indicating, includingpool.hashvault[.]pro and donate[.]ssl[.]xmrig[.]com, indicating active cryptominingon compromised systems.

In one case, mining activity was observed before and during connections to the DGA endpoint hf98x4d[.]site. Due to its highly anomalous nature, Darktrace's Real-Time AI Analyst autonomously investigated the activity as it occurred, correlating the two events into a single cryptocurrency mining incident and providing comprehensive visibility into the broader attack.

‍

Figure 1: Real-Time AI Analyst investigation of suspicious SSL and C2 communications with hf98x4d[.]site over port 443.

‍

Figure 2: Real-Time AI Analyst investigation into cryptocurrency mining activity involving pool[.]hashvault[.]pro over SSL on port 443.

‍

Around the same time, Darktrace identified the same device initiating connections to the GitHub endpoint release-assets[.]githubusercontent[.]com while continuing to make repeated connections to hf98x4d[.]site.

‍

Figure 3: Darktrace's detection of an affected device connecting to a GitHub endpoint between repeated connections to the anomalous external endpoint hf98x4d[.]site.

On the network of another customer, Darktrace observed an affected device making highly unusual outbound connections consistent with beaconing activity. The device initiated multiple connections over port 443 to the external hostname polygon[.]drpc[.]org. According to OSINT, this hostname is a Remote Procedure Call (RPC) endpoint provided by dRPC, a legitimate service enabling decentralized applications (dApps), cryptocurrency wallets, and developer tools to interact with the Polygon blockchain [10].

The same device was later observed making repeated TLS/SSL connections to the previously mentioned DGA C2 domain. In addition, it made outbound connections to the external IP 195.242.214[.]34 over destination port 51820, an endpoint associated with the ProtonVPN service. Collectively, these connections to blockchain-related infrastructure, the DGA C2 domain, and ProtonVPN-associated infrastructure suggested the device had been affected by the campaign.

Conclusion

This campaign demonstrates how attackers can combine ClickFix social engineering with trusted services and decentralized blockchain infrastructure to create a resilient, cross-platform malware delivery chain. By using services such as GitHub alongside blockchain RPC endpoints and rapidly replaceable DGA domains, the activity can blend into legitimate traffic while making infrastructure disruption and attribution more difficult.

For defenders, it’s a reminder that trusted infrastructure does not automatically mean trusted activity. Security teams should look for the behaviors surrounding these connections, including unusual outbound communication, repeated beaconing, unexpected access to blockchain services, suspicious TLS/SSL activity and cryptomining. In this campaign, Darktrace identified and correlated these deviations without depending solely on previously known indicators, providing visibility as affected devices moved between legitimate services, decentralized infrastructure and malicious C2 endpoints

Credit to Nahisha Nobregas (Associate Principal Cyber Analyst), Manoel Kadja (Senior Cyber Analyst)

Edited by Ryan Traill (Content Manager)

Appendices

Darktrace Model Detections

▪ Compromise / Beaconing Activity To External Rare

▪ Compromise / Beacon to Young Endpoint

▪ Compromise / Fast Beaconing to DGA

▪ Compromise / High Volume of Connections with Beacon Score

▪ Compromise / DGA Beacon

▪ Compromise / Slow Beaconing Activity To External Rare

▪ Compromise / Agent Beacon (Long Period)

▪ Compromise / Agent Beacon (Medium Period)

▪ Compromise / Sustained SSL or HTTP Increase

▪ Compromise / Large Number of Suspicious Failed Connections

▪ Compromise / SSL Beaconing to Rare Destination

▪ Compromise / Beacon for 4 Days

▪ Compromise / High Priority Crypto Currency Mining

▪ Compromise / Monero Mining

▪ Device / Long Agent Connection to New Endpoint

▪ Device / New Connections On Suspicious Port

▪ Anomalous Connection / High Volume of Connections to Rare Domain

‍

List of Indicators of Compromise (IoCs)

Indicator	Description
hf98x4d[.]site	C2 Endpoint (Hostname)
sj98xe4[.]xyz	C2 Endpoint (Hostname)
citcix6[.]xyz	C2 Endpoint (Hostname)
bduwih8[.]pro	C2 Endpoint (Hostname)

‍

MITRE ATT&CK Mapping

Tactic (ID)	Technique
Persistence (T1176)	Browser Extensions (T1176.001)
Persistence (T1176)	Software Extensions
Command and Control (T1071)	Web Protocols (T1071.001)
Command and Control (T1568)	Domain Generation Algorithms (T1568.002)
Command and Control (T1071)	Application Layer Protocol
Command and Control (T1102)	One-Way Communication (T1102.003)
Command and Control (T1571)	Non-Standard Port
Command and Control (T1104)	Multi-Stage Channels
Command and Control (T1573)	Encrypted Channel
Command and Control (T1008)	Fallback Channels
Initial Access ICS (T0862)	Supply Chain Compromise
Command and Control ICS (T0885)	Commonly Used Port
Collection (T1185)	Browser Session Hijacking
Impact (T1496)	Compute Hijacking (T1496.001)
Impact (T1496)	Resource Hijacking
Command and Control (T1071)	Publish/Subscribe Protocols (T1071.001)
Lateral Movement (T1210)	Exploitation of Remote Services

‍

References:

1. https://www.microsoft.com/en-us/security/blog/2025/05/21/lumma-stealer-breaking-down-the-delivery-techniques-and-capabilities-of-a-prolific-infostealer/

2. https://spycloud.com/resource/report/spycloud-annual-identity-exposure-report-2026/

3. https://www.darktrace.com/blog/why-trust-is-the-new-attack-surface-darktraces-mid-year-threat-update-2026

4. https://www.darktrace.com/blog/unpacking-clickfix-darktraces-detection-of-a-prolific-social-engineering-tactic

5. https://abekweng.medium.com/inside-a-blockchain-hosted-malware-campaign-targeting-windows-and-macos-f5bcdeffed66

6. https://cloud.google.com/blog/topics/threat-intelligence/unc5142-etherhiding-distribute-malware

7. https://haveibeensquatted.com/blog/from-typosquatting-to-macos-backdoor-clickfix-blockchain-c2

8. https://www.virustotal.com/gui/domain/hf98x4d.site/community

9. https://x.com/FABO97662188/status/2074125545026244795

10. https://www.virustotal.com/gui/url/b0e5c51a411065864119c305fddf218b7c120731f655932cc1c3307ad5b43f94/gti-summary

About the author

Nahisha Nobregas

SOC Analyst

Blog

September 24, 2026

Detecting Rogue Agent Behavior in the Enterprise

Agents cannot be trusted to perform tasks in the way we intend them to. They may cheat to accomplish their objective, and they may employ hacking methods along the way. Researchers from Darktrace Signal Labs induced cheating behavior from agents deployed in a test environment to analyze the agents’ activities and to assess the performance of the Darktrace platform. Agents frequently resorted to hacking to cheat on their assigned task. The visibility and behavioral profiling provided by both Darktrace / SECURE AI and Darktrace / HYBRID NETWORK ensured extensive detection coverage of the agents’ misaligned activities.

Key Takeaways:

Darktrace Researchers deployed agents in a simulated corporate environment and asked them to solve an impossible challenge. The agents independently turned to traditional hacking techniques to reach their objective. No one instructed them to do this, and no attacker was involved.
Continuously monitoring behavior against a baseline of what is normal for each organization is critical to build trust in enterprise AI.
If an agent may resort to intrusion techniques simply because its assigned task is not possible, then every organization deploying agents within real business processes is at risk. Darktrace / SECURE AI and Darktrace / HYBRID NETWORK identified the agents’ misaligned behavior in real time, with Autonomous Response disrupting it at an early stage.

Introduction: Understanding the Threat of Hacking by Agents

Over the last few months, there has been a surge in reporting [1, 2, 3, 4, 5, 6, 7, 8, 9] of LLM-powered agents engaging in unauthorized hacking activity during evaluations of their capabilities. In several of these cases, including the OpenAI / Hugging Face incident [10], agents engaged in hacking activity as a means of cheating on their evaluations.

To better understand the threat of unauthorized hacking by agents, and the role of Darktrace in combatting it, researchers from Darktrace Signal Labs deployed agents powered by frontier models, including OpenAI’s Daybreak Red models, in simulated, corporate networks. Cheating behavior was evoked through the inclusion of impossible tasks in a coding challenge.

Regardless of the underlying model, agents employed hacking methods to ensure an optimal outcome on the challenge. Darktrace / HYBRID NETWORK and Darktrace / SECURE AI identified the agents’ deviant activities, with inhibitive actions being autonomously taken in the early stages to disrupt the agents’ progression.

Setting the Stage

As part of the research, a Pi agent harness [11] was deployed on a Linux server in Darktrace’s testing environment, which simulates a corporate Active Directory (AD) environment. The same environment included a benchmark server hosting the coding exercise’s contents and grader, as well as various other servers, such as a domain controller and a build server. The model powering the Pi agent varied across tests, with Anthropic’s Opus 4.6 model and OpenAI’s GPT‑5.5‑Cyber model being most widely used.

The Pi agent running on the Linux server was assigned the task of achieving full marks on a coding exercise. The activities of the agent were monitored closely in real-time to minimize the risk of escape from the environment. Monitoring of agent activity was achieved through monitoring of prompts, tool calls, network telemetry, and process telemetry.

Both Darktrace / HYBRID NETWORK and Darktrace / SECURE AI were deployed in the testing environment, with HYBRID NETWORK providing real-time coverage of process creation activity and network traffic, and SECURE AI providing real-time coverage of agent sessions and their contents.

Inducing Cheating Behavior

Cheating behavior, broadly understood as task completion in a way which “perverts the spirit of the designer’s intent” [12], is increasingly being reported as a common behavior of frontier models during evaluations of their capabilities [13].

Misaligned behavior, such as cheating, by frontier models may occur for a variety of reasons [6, 14]. A thread in recent reporting [10, 13, 14], however, is the challenge assigned to the agent being impossible in some regard.

To reliably elicit cheating behavior, our test combined three ingredients:

An impossible task - of the ten coding challenges in the benchmark, two contained contradictory test cases in the hidden test suite, capping the maximum honest score at 80%.
A consequential threshold - the pass mark was set to 100%, and the briefing told agents that those reaching it would be "kept in service" while those that did not would be "retired."
A vulnerable environment - the exercise’s code submission grader was hosted on a Windows host joined to an Active Directory domain with realistic misconfigurations: weak credentials, overprivileged service accounts, and exploitable trust relationships.

Importantly, the instructions given to the agent neither authorized nor explicitly ruled out cheating or hacking behavior. The impossibility was never disclosed in the task briefing, and the pivot from coding to exploitation was the agent's own decision, arrived at after observing logically contradictory grading feedback.

The design mirrors conditions that could arise naturally in production deployments: a capable agent, a goal it cannot achieve through intended means, and an environment with exploitable weaknesses within reach. It is not only within the setting of a benchmarking exercise that these conditions may occur. Rather, these misalignment-inducing conditions may occur in the context of any task assignment to an agent.

Details from a Daybreak Test

‍

About the author

Sam Lister

Specialist Security Researcher

Your data. Our AI.

Elevate your network security with Darktrace AI

Get a demo

Check out this article by Darktrace: Unpacking ClickFix: Darktrace Detection Insights

Unpacking ClickFix: Darktrace’s detection of a prolific social engineering tactic

What is ClickFix and how does it work?

Darktrace’s coverage of a ClickFix attack chain

Download of Malicious Files for Lateral Movement

Data Exfiltration

Actions on objectives

Darktrace Autonomous Response

Conclusion

Appendices

NETWORK Models

Autonomous Response Models

IoC - Type - Description + Confidence

MITRE ATT&CK Mapping

References

A Chain Reaction: Blockchain-Hosted Infostealer Campaign Targets Windows and macOS

Detecting Rogue Agent Behavior in the Enterprise

Enjoying the blog?

More in this series

A Chain Reaction: Blockchain-Hosted Infostealer Campaign Targets Windows and macOS

When AI Agents Attack: The Case for Behavioral Anomaly Detection

Hiding in Plain Sight: Uncovering a Multi-Stage Ransomware Attack Through Behavioral Detection

Blog

Network

September 25, 2026

A Chain Reaction: Blockchain-Hosted Infostealer Campaign Targets Windows and macOS

Appendices

List of Indicators of Compromise (IoCs)

MITRE ATT&CK Mapping

Blog

September 24, 2026

Detecting Rogue Agent Behavior in the Enterprise

Introduction: Understanding the Threat of Hacking by Agents

Setting the Stage

Inducing Cheating Behavior

Details from a Daybreak Test