Skip to content

Image: mkd-cirt.mk · rights & removal

Executive Summary

A sophisticated malware distribution framework known as ROMULUS ClickFix has been identified, facilitating the spread of previously undocumented Remote Access Trojans (RATs), including variants of CNCMachineRMS and Legion Loader. This malware utilizes ClickFix techniques, involving deceptive messaging and the abuse of legitimately signed executables, to deliver malicious code. Attacks have leveraged customized GPT tools on platforms like ChatGPT to redirect unsuspecting users toward compromised websites. Once directed, users encounter ClickFix scams that employ DLL sideloading, often through files such as MSI installers, to execute shellcode that activates the controlling RAT.
The threat extends beyond ROMULUS; similar ClickFix methodologies have been observed in multiple other attack campaigns, including those involving the EtherHiding technique, fake review pages, compromised WordPress installs, and abusing Google Ads. These diverse infection vectors demonstrate a pattern of attacker adaptability. The resulting RATs grant attackers extensive control over compromised workstations, enabling surveillance, file access, and data theft, with communications routed via DNS over HTTPS (DoH).

Facts Only

Hacker actors utilized ROMULUS ClickFix to distribute a previously undocumented Remote Access Trojan (RAT) similar to CNCMachineRMS.
Blackpoint Cyber researchers documented that ROMULUS likely functions as a distribution service for CNCMachineRMS and Legion Loader via separate attack chains.
ROMULUS was first publicly documented last month by Forescout Research as a malware distribution service employing deceptive messages and ClickFix instructions.
Customized GPT tools on ChatGPT were used to redirect users to malicious websites that deliver RAT malware via DLL sideloading.
A malicious MSI installer, "IconEdit2Turb.msi," executed via "msiexec.exe" with a PowerShell command, initiating a DLL sideloading chain to activate CNCMachineRMS.
The RAT allows remote control, live screen image transfer, microphone/camera recording, file management, and system information collection.
The Trojan communicates with attackers' infrastructure using DNS over HTTPS (DoH).
Separate campaigns documented utilize ClickFix alongside EtherHiding, fake Zillow review pages, the TRUST#GRAFT attack chain on WordPress, and abusing sponsored Google Search results.
Attackers also leverage Google Ads to redirect users to fake download pages containing ClickFix commands.
The malware has been observed in complex chains enabling persistent presence, internal spread, and login data exfiltration.

Full Take

The prevailing narrative suggests a sophisticated evolution in initial access vectors: the malware itself is merely the payload, while the *delivery mechanism* is what has become highly variegated and resilient. The consistent thread linking these disparate attacks is the application of ClickFix-like deceptive persuasion layered onto increasingly trusted or novel entry points—from customized LLMs on ChatGPT to legitimate Google Ads placements. This signals a high degree of resource investment in operational security and social engineering ingenuity by the threat actors.
The pattern observed is not just "malware spreading," but rather "delivery vector polymorphism," where the core command sequence (e.g., DLL sideloading activated by a trusted process) remains stable while the entry point morphs. This adaptability challenges traditional signature-based defense models. Furthermore, the convergence of various techniques—from DNS over HTTPS for C2 exfiltration to steganography for payload reconstruction—points to a strategic aim beyond simple opportunistic theft; it suggests a focus on achieving persistent, low-profile presence within targeted environments.
The primary assumption driving this threat model is that user trust in digital interfaces (search results, AI assistants, system installers) is a quantifiable, exploitable vulnerability. The cost bearer, initially, is the end-user, but the secondary beneficiaries are potentially state-affiliated or highly organized criminal enterprises capable of maintaining such complex, multi-stage operations across disparate platforms.
What perspectives are missing? How does the adoption of legitimate services (like Microsoft executables or Google infrastructure) as attack conduits change the burden of proof for defenders? What research is currently tracking the defensive resilience of third-party LLM implementations against prompt injection leading to redirection?
Patterns detected: none
Autonomy check: No autonomy signals.

From the original · North Macedonia MKD-CIRT Alerts

ROMULUS ClickFix attacks spread a RAT malware that can spy on users and take control of computers Published: 20.11.2026 Attacks have been recorded in which hackers used the ROMULUS ClickFix malware distribution platform to spread a previously undocumented Remote Access Trojan (RAT) that shares similarities with CNCMachineRMS.
Read the full story at mkd-cirt.mk

Sentinel unavailable

The automated check of the source article's wording did not complete, so no result is shown.

This looks only at the wording of the original source article, not at this page's AI-written sections. A small local AI model made this estimate. It has not been checked against known human and machine texts, so treat it as provisional. It cannot show who wrote an article.

The attack with ROMULUS ClickFix is a malicious software that can spy on users and take control of computers. | Huntaegis