The 2026 Vishing Offensive: Com-Affiliated Threat Actors Unleash Data Heists
In a significant escalation of cyber threats, Com-affiliated threat actors have orchestrated a series of high-profile heists throughout 2026, primarily leveraging sophisticated vishing (voice phishing) attacks. Palo Alto Networks Unit 42’s latest analysis reveals persistent campaigns targeting data-rich industries such as financial services, healthcare, and legal sectors. These actors, including named groups like Bling Libra (aka ShinyHunters) and newly identified clusters such as CL-CRI-1264 (Vortex) and CL-CRI-1274 (Cinder), have successfully executed data theft and extortion schemes, predominantly impacting organizations in the U.S., UK, and Australia.
Table of Contents
The observed activity, spanning from August 2026 into early 2027, highlights a strategic shift towards exploiting human trust through social engineering. The attackers’ methods are designed to bypass traditional security measures by directly engaging with employees, making these operations particularly insidious. This report delves into the evolving tactics, techniques, and procedures (TTPs) employed by these Com-affiliated threat actors.
Emerging Attack Vectors and Infrastructure
The Vortex and Cinder Clusters: New Fronts in the Cyber War
September 2026 marked the emergence of two new activity clusters, CL-CRI-1264 (Vortex) and CL-CRI-1274 (Cinder), both exhibiting distinct yet related operational patterns. The Vortex cluster notably focused on healthcare and pharmaceutical organizations, employing vishing to impersonate internal IT support. Attackers directed employees to adversary-in-the-middle (AiTM) proxy servers, enabling real-time capture of credentials and multi-factor authentication (MFA) approvals.
These actors demonstrated alarming agility, enrolling unauthorized MFA devices within minutes of initial authentication. A recurring observation was the reuse of a specific device, the Pixel 10a, across compromised accounts. Data exfiltration was automated using Python scripts targeting Microsoft 365 mailboxes, OneDrive, and SharePoint. Post-exfiltration, victims received extortion messages via Microsoft Teams, directing them to Tor-hosted negotiation sites.
Cinder’s Sophisticated Session Hijacking
Similarly, the Cinder cluster utilized vishing for initial access but distinguished itself by replaying persistent “keep me signed in” session tokens to bypass Duo MFA. This allowed attackers to maintain access through proxy infrastructure. They leveraged scripted access via python-requests and the Microsoft Graph SDK to exfiltrate data from SharePoint and OneDrive.
Extortion demands were disseminated internally through email and Microsoft Teams, again guiding victims to a Tor-based negotiation platform. The consistent use of cloud collaboration platforms and SSO by targeted organizations facilitates these rapid, widespread attacks.
Malicious Browser Extensions: A Stealthy Infiltration Tool
The “Google Chrome” Masquerade
In investigations linked to CL-CRI-1192 (also tracked by Google as UNC6671), Unit 42 uncovered a malicious browser extension deceptively named “Google Chrome.” This extension, not available on the official Chrome Web Store, was downloaded from an attacker-controlled domain, oskeysetup[.]com, requiring manual sideloading via Developer Mode. This naming convention cleverly masked the browser’s built-in warning banner during debugging.
The extension’s capabilities are extensive, including global keylogging, theft of local/session storage and IndexedDB data, harvesting of authentication cookies for Microsoft and Google accounts, and scanning for QR codes to capture login or MFA tokens. It even features a command-and-control (C2) module that allows remote control via the Chrome DevTools Protocol.
This module enables operators to capture hidden screenshots, dispatch input events, steal clipboard contents, and perform remote site navigation. A second version of this extension (1.6.112) was identified, communicating with the same C2 server and sharing a hard-coded API key. This suggests a deliberate effort to create a persistent, versatile credential-harvesting tool for Chromium-based browsers.
Victimology and Strategic Targeting
Industries and Geography Under Fire
The operational tempo of Com-affiliated threat actors accelerated significantly around August 2026. Victims consistently possess large volumes of sensitive corporate data, where exposure carries severe regulatory, legal, or market consequences. Financial services, healthcare, and legal organizations remain primary targets, though high technology, construction, and transportation sectors are also affected.
Attackers systematically target individuals or systems responsible for high-value files, documents, and databases. This includes partner communications, firm strategy documents, pending contracts, private investment portfolios, and proprietary research data. The threat actors explicitly leverage the potential for reputational damage and regulatory penalties in their extortion tactics.
Geographically, the U.S. has been the most targeted nation. However, a notable increase in targeting Australian organizations began in late September 2026, with Australia becoming the second most targeted country by October. This geographical shift may indicate an expansion of the threat actors’ operational reach or a strategic focus on specific regulatory environments.
Tactics, Techniques, and Procedures (TTPs) Deep Dive
Infrastructure and Tooling: The AiTM Playbook
A hallmark of Com-affiliated threat actors is their use of apex domains impersonating SSO portals, IT support pages, and passkey configuration services. These domains host short-lived subdomains that mimic target organizations, such as {company}.mfaregister[.]com. Unit 42 actively tracks this infrastructure and shares findings through its GitHub repository.
Similarities in hosting infrastructure, attack chains, and victimology suggest potential collaboration or shared tooling among these groups. The Doko’s Panel, an AiTM phishing kit, has been observed, with some variants replacing the default backend handler with a renamed endpoint, api_FyekIDWY.php, which has been identified on multiple Com-affiliated domains. This indicates a standardized, modular approach to their attack infrastructure.
The observed AiTM sequences, from credential entry to MFA number-match prompts, strongly indicate successful session captures. This sophisticated approach, combined with malicious browser extensions and targeted social engineering, presents a formidable challenge. As we move into 2027, the adaptability and persistence of these Com-affiliated threat actors underscore the critical need for enhanced cybersecurity defenses and continuous threat intelligence monitoring.
Source: Palo Alto Networks Unit 42 | unit42.paloaltonetworks.com


