CyberSecurity SEE

XaocMail Launches ClickFix Campaign with Phantom53 C2 Deployment

XaocMail Launches ClickFix Campaign with Phantom53 C2 Deployment

Security researchers from SOCRadar have recently revealed a sophisticated phishing campaign orchestrated by Russian-speaking cybercriminals. This operation leverages artificial intelligence (AI) to enhance its infrastructure and efficiency, indicating a notable evolution in the tactics employed by threat actors.

The campaign, which has reportedly been in motion since August 2026, was preceded by nearly a year of preparatory groundwork that began in October 2025. It utilizes an intricate array of tools, including XaocMail, a self-hosted phishing platform, Phantom53, a stealthy command-and-control (C2) framework, and Havoc, a post-exploitation toolkit. These components work in unison to execute a highly effective phishing strategy.

The initial assault begins with XaocMail, which crafts personalized phishing emails that masquerade as communications from reputable document-sharing services. Targets are lured into believing the emails are from well-known platforms such as Google Docs, Adobe Acrobat, Office365, and SecuredDocs. To facilitate this deception, the emails include links to Cloudflare Workers pages, employing ClickFix-style lures. These lures employ social engineering techniques that encourage the victims to execute commands through the Windows Run dialog (Win+R).

Recent versions of this phishing scheme have escalated in sophistication, delivering MSI installer files that imitate legitimate software from reputable vendors including Microsoft, Python, and Node.js. Each phishing email contains a unique link identifier, enabling the attackers to log critical telemetry data around the geographic and network origins of the victims when they engage with the links.

Phantom53 acts as the backbone of this operation, serving as the command-and-control framework responsible for managing communications between the attackers and compromised machines. It operates via encrypted DNS-over-HTTPS queries directed at public resolvers such as Google, Cloudflare, and ControlD. This method conceals malicious traffic within encrypted sessions, making it exceedingly challenging for security tools to detect any nefarious activity. The framework is architected in three distinct layers: a lightweight beacon that polls for new commands every twelve minutes, a watcher component that keeps an eye out for triggers from the operators every three minutes, and the Havoc Demon agent, which is deployed on demand for targets of high importance. To avoid detection, all components are executed within isolated Python virtual environments utilizing signed pythonw.exe processes. The framework even includes a domain generation algorithm to ensure continuity, allowing it to employ alternative command-and-control domains if its current ones face blocks or disruptions.

The implications of this phishing campaign are extensive, as it targets organizations across diverse sectors such as technology, finance, education, government, logistics, and manufacturing in regions including Latin America, Asia, and Europe. By relying entirely on legitimate cloud services, the attackers use Cloudflare Workers for the delivery of payloads, Cloudflare KV for data storage, and public DNS-over-HTTPS resolvers for their command-and-control communications.

Analysis of the campaign’s coding structure suggests that the threat actors have employed AI-assisted development tools—specifically utilizing Claude Code task structures and automated testing frameworks. This indicates a conscious effort to streamline their development cycle and enhance their ability to evade security measures.

To mitigate the risks posed by such an advanced campaign, organizations are urged to adopt several precautionary measures. Security teams should remain vigilant and monitor for scheduled tasks that mimic legitimate Microsoft, Adobe, or OneDrive services, particularly those featuring random alphanumeric suffixes. Additionally, inspecting DNS-over-HTTPS traffic patterns for abnormal polling intervals—especially connections every twelve or three minutes to public resolvers—can help identify potential compromise.

Moreover, email security controls should be configured to flag document-sharing notifications containing links to Cloudflare Workers domains. Enhanced endpoint detection measures should alert administrators to any unauthorized installations of MSI files that introduce Python runtimes into user directories. Furthermore, network defenders are advised to restrict or monitor DNS-over-HTTPS traffic and implement robust application control policies to prevent the execution of unsigned MSI files without administrative consent.

In light of these findings, awareness and proactive measures remain critical in combating the rising tide of phishing attacks propelled by advanced technological tactics. Organizations across the globe must deepen their defensive strategies to stay ahead of increasingly sophisticated threats.

Source link

Exit mobile version