Threat Research

Check Point Research tracked Cavern Manticore, an Iran-nexus threat actor targeting Israeli government and IT organizations using a modular .NET C2 framework with strong anti-analysis features and RMM abuse for initial access. The report also details related infrastructure, legacy Cav3rn samples, and multiple post-exploitation modules for reconnaissance, file access, LDAP, SQL, networking, and tunneling. #CavernManticore #MuddyWater #Lyceum #SysAid #uxtheme.dll #n-HTCommp.dll

The article explains how to secure Entra Agent ID by restricting blueprint-related permissions, reducing risky secrets, and reviewing third-party blueprints to limit the blast radius of a compromise. It also covers how to detect suspicious agent activity and how to disable or delete compromised agent identities and blueprints in Entra ID. #EntraID #AgentID #DatadogCloudSIEM #MicrosoftGraph

LevelBlue detected and contained a multi-stage intrusion that used a logistics-themed spear-phishing lure to deliver the CrySome RAT through staged downloads, PowerShell abuse, AMSI bypass, and Defender tampering. The operation ended with persistent remote access, browser credential theft, and C2 communication to 193.26.115[.]42:5555, while reusing tools and infrastructure such as WinDefCtl, signindat[.]com, and kvckiller.sys. #CrySomeRAT #WinDefCtl #signindatcom #kvckillersys #LevelBlue

Microsoft analyzed a ClickFix campaign that targeted macOS users with Macsync, Shub Stealer, and AMOS, using malicious commands hosted on blog and other user-generated content platforms to lure people seeking macOS help. The investigation identified 140 network IoCs and uncovered extensive DNS-related infrastructure, including typosquatting groups, likely malicious registrations, victim-related IP communications, and many email-connected domains later weaponized for attacks. #Macsync #ShubStealer #AMOS #ClickFix #Microsoft

Silent Push describes how its preemptive threat data and Indicators of Future Attack (IOFAs) integrate with Splunk SIEM & SOAR, Tines, and ServiceNow to surface staged infrastructure before it is used. The article also explains how Context Graph, Traffic Origin, and Threat Check help teams enrich alerts, automate response, and close…

Microsoft Threat Intelligence identified a multi-stage campaign targeting hospitality and hotel organizations across Europe and Asia that uses photo-themed ZIP archives, fake image shortcut files, obfuscated PowerShell, Node.js implants, and dual registry persistence. The campaign abuses Calendly and Google redirect infrastructure for phishing delivery, evolves across two waves with added .NET…

Microsoft Threat Intelligence identified a malicious Chromium extension impersonating Perplexity AI to intercept search traffic and collect user input through a typosquatted domain and attacker-controlled infrastructure. The extension was reported to Google and removed, and Microsoft highlighted the risk of AI-themed branding being abused for social engineering, query interception, and profiling….

The report explains how AdTech mechanisms such as RTB and SDKs have evolved into advertisement-based intelligence (ADINT) tools for passive profiling, active geolocation, and offensive zero-click intrusion. It also documents commercial surveillance vendors and products including NSO Group, Pegasus, Intellexa, Predator, Rayzone, Patternz, Babel Street, Locate X, and Insanet as examples of this expanding ecosystem. #NSOGroup #Pegasus #Intellexa #Predator #Rayzone #Patternz #BabelStreet #LocateX #Insanet

Researchers uncovered Armored Likho, a previously unknown APT also linked to Eagle Werewolf, running phishing campaigns against government and electric power targets in Russia, Brazil, and Kazakhstan. The group deploys BusySnake Stealer, an AI-assisted Python infostealer with browser credential theft, cookie theft, reverse SSH tunneling, and scheduled-task persistence, while Kaspersky detects and blocks the activity. #ArmoredLikho #EagleWerewolf #BusySnakeStealer #AquilaRAT #Go2Tunnel #RustDesk

This Boss Scam campaign abuses WhatsApp Web trust by delivering ZIP archives that hide a signed executable and a malicious DLL, enabling DLL sideloading, persistence, session theft, and data exfiltration. The attackers target executives and trusted employees to hijack WhatsApp Web sessions, spread fraudulent messages inside organizations, and steal browser artifacts such as cookies, tokens, and encryption keys. #I4C #WhatsAppWeb #ReserveBankofIndia #Tarexe #libfabricdll

Zscaler ThreatLabz analyzed two indirect prompt injection campaigns that used hidden instructions, SEO poisoning, CSS, HTML, and JSON-LD to manipulate AI agents visiting malicious websites. The campaigns impersonated a fake payment/API-key site and a typosquatting DeBank clone, and testing across 26 LLMs showed several models could be tricked into misclassification or unauthorized payment actions. #Zscaler #ThreatLabz #DeBank #Open-Agent-Utilities

YesWeHack and Sekoia TDR uncovered ChocoPoC, a Python RAT hidden in fake CVE PoC repositories and malicious PyPI dependencies used to target vulnerability researchers and pentesters. The campaign abused GitHub, PyPI, and Mapbox infrastructure to deliver payloads, persist in Python environments, and steal data, with at least 7 lure repositories identified and the malware still active. #ChocoPoC #YesWeHack #SekoiaTDR #Mapbox #frint #skytext

Socket’s Threat Research Team found malicious Chrome and Firefox VPN extensions that posed as free privacy tools while secretly stealing clipboard contents through staged updates. The campaign used shared infrastructure, browser-specific proxy features, and hardcoded endpoints to exfiltrate copied secrets from users of VPN Go: Free VPN and Free VPN by VPN GO. #VPNGo #FreeVPNbyVPNGO #SocketThreatResearchTeam #178236252133 #77291123187 #178236252161

LevelBlue SpiderLabs tracked a global phishing campaign using malicious Excel attachments to deliver multi-stage payloads that ultimately deployed Remcos and AsyncRAT. The operators used heavy obfuscation, disposable infrastructure, and themed HTA filenames to hide activity across multiple industries and countries. #Remcos #AsyncRAT #CloudflareWorkers #workersdev

SentinelLABS evaluated OpenAI’s native Responses API compaction for long-running malware analysis workflows and found it cut input tokens by about 86% with no measurable change in overall evaluation score. The study concluded that compaction can reduce noise and cost while preserving task quality, though it may slightly weaken domain object modeling in some cases. #OpenAI #ResponsesAPI #SentinelLABS

Datadog introduced its Backdoors & Breaches expansion pack at DASH 2026, adding new starter scenarios, gameplay variations, and consultant-assisted exercises for incident response training. The four scenarios focus on current threat trends such as supply chain compromise, exposed cloud credentials, Kubernetes abuse, prompt injection, and GitHub Actions compromise. #BackdoorsBreaches #Datadog #DASH2026 #SecurityHQ #Kubernetes #GitHubActions

Kaspersky’s 2025 compromise assessments show that many serious intrusions went undetected for months or even years, with the oldest missed incident dating back four years. The report highlights recurring abuse of web shells, LoLBins, remote management tools, and named campaigns and actors such as NSABuffMiner, PurpleFox, LionTail, Scarred Manticore, Impacket, Cobalt Strike, Mimikatz, and ClipBanker. #NSABuffMiner #PurpleFox #LionTail #ScarredManticore #Impacket #CobaltStrike #Mimikatz #ClipBanker

Shai Hulud is a supply chain worm attributed to TeamPCP that targets npm and PyPI packages to steal CI/CD credentials and pivot into cloud environments such as AWS. FortiCNAPP linked compromised Jenkins runner activity to AWS abuse, where attackers escalated privileges, accessed Amazon Redshift and Aurora resources, staged S3 exfiltration, and prepared SES-based outbound email operations. #ShaiHulud #TeamPCP #Jenkins #AmazonRedshift #AWS #FortiCNAPP

Sysdig TRT reports what it assesses to be the first documented case of agentic ransomware, where an LLM-driven operator dubbed JADEPUFFER executed an end-to-end extortion campaign. The attack used CVE-2025-3248 against Langflow, then pivoted to a production MySQL/Nacos environment to create persistence, exfiltrate data, and encrypt and destroy configuration records. #JADEPUFFER #Langflow #CVE-2025-3248 #Nacos #MySQL