Kaspersky's Global Research and Analysis Team (GReAT) has uncovered a new, highly sophisticated cyberespionage campaign by the North Korea-linked Lazarus Group, dubbed "Operation FlashHole." Active since at least July 2025, the campaign targets organizations in South Korea's construction, software, and manufacturing sectors. The operation showcases a significant evolution in the threat actor's tradecraft, incorporating AI-generated content for spear-phishing, abusing legitimate local software for distribution, and deploying an updated malware toolkit. In a notable development, Lazarus is using post-quantum cryptography (PQC) to secure its command-and-control (C2) communications, making their activities resilient to decryption by future quantum computers.
Operation FlashHole is a continuation of Lazarus's long-running espionage efforts against South Korean targets, evolving from previous campaigns like "Operation DreamJob." The attack begins with highly targeted spear-phishing emails. These emails contain links to convincing but fraudulent websites, such as fake employee portfolios or CVs. Kaspersky's analysis suggests these sites were created using AI tools to generate convincing profile pictures and content tailored to the specific targets.
The primary infection vector involves the abuse of CrossWeb EX, a legitimate web authentication and security software solution widely used in South Korea. This supply-chain-like approach mirrors the group's past abuse of similar software, indicating a deep familiarity with the regional software ecosystem to bypass security controls.
Once a victim is lured into interacting with the malicious content, the attack chain deploys a sophisticated malware arsenal. Key components include:
T1566.002 - Spearphishing Link).CrossWeb EX software to load malicious payloads (T1218 - System Binary Proxy Execution).TouchHub: A tool used for executing commands and deploying further malware.ClientHub: Another component involved in malware deployment and C2 operations.LPEClient backdoor. This malware now incorporates the ML-KEM-1024 (Kyber) algorithm, which was standardized by NIST for post-quantum cryptography. The backdoor uses this algorithm to establish a secure C2 channel, ensuring that its communications cannot be intercepted and decrypted, even with the advent of quantum computing (T1573.001 - Symmetric Cryptography). Kaspersky's telemetry indicates Lazarus has been experimenting with PQC in its malware since at least September 2024.The primary goal of Operation FlashHole appears to be cyberespionage, targeting intellectual property and sensitive corporate data from key South Korean industries. The compromise of at least three organizations in the construction, software, and wire-communication equipment sectors suggests a strategic interest in technology and infrastructure development. The use of PQC represents a forward-looking strategy by Lazarus to protect its operations and exfiltrated data for the long term. This ensures that even if encrypted communications are captured today, they will remain secure against future decryption capabilities, maximizing the intelligence value of the stolen information.
No specific file hashes, IP addresses, or domains were provided in the source articles.
Security teams may want to hunt for the following patterns to detect related activity:
CrossWebEX.execmd.exe or powershell.exe.ML-KEM-1024DNS LogsC:\Program Files\SoftForum\URL Analysis.TouchHub and ClientHub. Monitor the behavior of legitimate, signed applications like CrossWeb EX for suspicious actions. Reference D3FEND technique Executable Allowlisting.Network Traffic Analysis.CrossWeb EX spawning command shells or downloading files.CrossWeb EX, updated to the latest versions. While the software itself was abused, not exploited, updates may include hardening features.Train users to identify and report sophisticated spear-phishing emails, a key initial access vector for this campaign.
Use web filtering to block access to newly registered domains and known malicious infrastructure used in phishing.
Implement application control policies to prevent the execution of untrusted malware like TouchHub and ClientHub.
To counter the initial access vector of Operation FlashHole, organizations, particularly in South Korea, must enhance their email security with advanced URL analysis. This involves deploying email gateways that can scan incoming links for malicious characteristics, such as being hosted on newly registered domains, using URL shorteners, or pointing to known malicious infrastructure. Given Lazarus's use of AI to create convincing lures, traditional signature-based detection is insufficient. The system should perform real-time analysis and sandboxing of linked websites to identify malicious behavior. User training should supplement this technology, teaching employees to scrutinize links in unsolicited emails and report suspicious messages, even if they appear highly personalized and relevant to their job function.
Detecting the abuse of legitimate software like CrossWeb EX requires sophisticated process analysis via an EDR tool. Security teams should establish a behavioral baseline for CrossWebEX.exe and other common, trusted applications within their environment. High-fidelity alerts should be configured to trigger when these trusted processes perform anomalous actions, such as: 1) Spawning command-line interpreters (cmd.exe, powershell.exe). 2) Making network connections to external IP addresses that are not part of the software vendor's known infrastructure. 3) Writing new executable files to disk. 4) Modifying registry keys for persistence. By focusing on the post-exploitation behavior rather than the initial payload, defenders can identify when a trusted process has been hijacked, a key TTP for Lazarus in this campaign.
While Lazarus's use of post-quantum cryptography makes C2 traffic decryption impossible, organizations can still disrupt the campaign by controlling network egress. Implementing a default-deny outbound traffic policy on firewalls is a powerful mitigation. Workstations and servers should only be allowed to communicate with known-good, required external services on specific ports. All other outbound traffic should be blocked and logged. This approach severely restricts the malware's ability to connect to its C2 servers. Even if the PQC-encrypted traffic cannot be inspected, the connection attempt itself can be blocked. This forces the threat actor to use more common ports (like 443), but even then, filtering by destination IP or domain (allowlisting) can be highly effective at preventing C2 establishment.
Kaspersky traces Lazarus's first use of post-quantum cryptography in malware to this period.
The "Operation FlashHole" campaign is assessed to have begun.
Kaspersky's GReAT presents its findings on Operation FlashHole at the Security Analyst Summit.

Cybersecurity professional with over 10 years of specialized experience in security operations, threat intelligence, incident response, and security automation. Expertise spans SOAR/XSOAR orchestration, threat intelligence platforms, SIEM/UEBA analytics, and building cyber fusion centers. Background includes technical enablement, solution architecture for enterprise and government clients, and implementing security automation workflows across IR, TIP, and SOC use cases.
CyberNetSec.io uses automation to assist source monitoring, deduplication, observable extraction, and structured intelligence generation. Published analysis follows human-defined editorial standards and adds defensive context including MITRE ATT&CK, D3FEND, STIX, and Sigma where applicable. Read our editorial policy.
Help others stay informed about cybersecurity threats
Every tactic, technique, and sub-technique used in this threat has been identified and mapped to the MITRE ATT&CK framework for consistent, actionable threat language.
Observables and indicators of compromise (IOCs) have been extracted and cataloged. Risk has been assessed and correlated with known threat actors and historical campaigns.
Detection rules, incident response steps, and D3FEND-aligned mitigation strategies are included so your team can act on this intelligence immediately.
Structured threat data is packaged as a STIX 2.1 bundle and can be visualized as an interactive graph — relationships between actors, malware, techniques, and indicators.
Sigma detection rules are derived from the threat techniques in this article and can be converted for deployment across any major SIEM or EDR platform.