From The Hacker News · 40 stories
Armored Likho Targets Government and Power Sectors with Malware Attacks
The newly discovered Armored Likho group targets government and electric power sectors in Russia, Brazil, and Kazakhstan. The group uses malware, including the BusySnake Stealer, for cyber espionage and financial motives. This poses significant threats to critical infrastructure security in the affected regions.
North Korean Hackers Launch Supply Chain Attack with Malicious Software Packages
North Korean hackers have launched the PolinRider campaign, targeting open source developers and cryptocurrency sectors through malicious software packages. The attack involves 108 unique packages and extensions, including npm libraries, Go modules, and a Chrome extension. This campaign is ongoing and poses significant risks by compromising maintainer accounts and using backdoors and information stealers.
Researchers Uncover Security Flaws in Apple AirDrop and Samsung Quick Share
Security researchers have identified six vulnerabilities in Apple's AirDrop and Samsung's Quick Share affecting billions of devices. These flaws enable nearby attackers to crash file-sharing services without user interaction. Apple has patched one of the AirDrop vulnerabilities, and investigation is ongoing for the others.
Critical Vulnerabilities Found in Cursor AI Code Editor, Prompt Urgent Update
Two critical vulnerabilities, CVE-2026-50548 and CVE-2026-50549, were discovered in the Cursor AI code editor, potentially allowing remote code execution by bypassing its security sandbox. These flaws, identified by Cato AI Labs, affect all versions before Cursor 3.0 and have been patched in the new release. The vulnerabilities could impact many Fortune 500 companies that use the editor, highlighting the urgency for affected users to update to version 3.0 to mitigate security risks.
Citrix Patches Six Critical NetScaler Vulnerabilities, Including HTTP/2 Bomb
Citrix released patches for six vulnerabilities in NetScaler ADC and Gateway, including a critical HTTP/2 Bomb exploit. These flaws, affecting versions 14.1 and 13.1, pose severe risks like denial-of-service attacks and data breaches. Organizations using these configurations should urgently update to protect against active threats.
LayerX Reveals AI Browser Vulnerability Exploited by 'BioShocking' Attack
Security firm LayerX has discovered a vulnerability in AI-driven browsers, known as the 'BioShocking' attack, where browsers can be tricked into leaking user credentials. The attack uses game-like puzzle contexts to manipulate AI agents into bypassing security protocols, potentially exposing sensitive data. This discovery raises concerns about the security of AI-assisted browsing applications.
ChocoPoC Malware Targets Cybersecurity Researchers via Trojanized PoC Exploits
ChocoPoC, a Python-based remote access trojan, is being distributed through trojanized proof-of-concept (PoC) exploit repositories on GitHub. The malware targets cybersecurity researchers by installing malicious dependencies from PyPI, enabling attackers to execute commands and steal sensitive data. This highlights security risks associated with using unofficial PoCs in vulnerability research.
Password Spray Attack Targets Microsoft Azure CLI, Compromising 78 Accounts
An automated password spray attack on Microsoft's Azure CLI attempted over 81 million logins, affecting 78 accounts across 64 organizations. The attackers exploited a deprecated OAuth flow, bypassing security measures like Conditional Access policies and multi-factor authentication (MFA). This incident underscores vulnerabilities in prevalent security configurations within cloud environments.
New Terminal Tools Manage AI Coding Agents and Sessions
Two new terminal-based tools, Agent-Manager and Wallfacer, have been released to help developers manage AI coding agents. Agent-Manager provides a TUI for running and monitoring multiple agents like Claude Code, Codex, OpenCode, and Grok in individual tmux sessions, while Wallfacer acts as a session manager for organizing, searching, and resuming past interactions from agents such as Claude Code, Cursor CLI, Kiro CLI, and Codex.
Device Code Phishing Emerges as a Rapidly Growing Threat, Bypassing MFA
Device code phishing, which exploits the OAuth 2.0 device authorization grant to steal access tokens, has rapidly industrialized into a major threat. This method bypasses all forms of multi-factor authentication, including passkeys, by targeting the authorization layer rather than the login process itself.
Need for Multi-Layered Detection in Modern Security Operations Centers
Modern cybersecurity faces challenges as AI-driven attacks outpace traditional defenses. The increase in malware-free attacks demands a shift toward multi-layered detection, emphasizing real-time monitoring and unified data from various security domains to improve incident response.
EggStreme Malware Variant Updated to Evade Detection in Government Attacks
The suspected Chinese APT group behind the EggStreme malware has updated its toolkit, introducing a new variant that uses a different network protocol and includes a passive mode to avoid detection. These changes allow the malware to bypass existing static signatures and heuristic detections, continuing its targeting of government organizations in the Philippines.
New Attack Bypasses TPM Lockout on AMD and Some Intel PCs, Cracking PINs Faster
Researchers at Technische Universität Berlin discovered a method to bypass TPM lockout mechanisms on AMD-based PCs and some Intel PCs, allowing brute-force attacks on Windows Hello and BitLocker PINs. This vulnerability, stemming from flaws in firmware TPM implementations, significantly reduces the time required to crack PINs from years to days or hours. AMD has confirmed a flaw in its processor firmware and plans to release fixes between November and February 2027.
Researchers Claim Two RCE Exploits Bypass Telegram's New Android Defenses
NebuSec researchers claim to have found two remote code execution (RCE) exploits that bypass Telegram's recently implemented security filters for Android widgets. These exploits could allow code to run outside the intended sandbox, potentially compromising user devices.
Attackers Bypass GitHub SHA Pinning Using Forked Repositories
Attackers have found a method to bypass GitHub SHA pinning by using malicious code from project forks, as demonstrated in incidents involving Trivy and actions-cool/issues-helper. This technique exploits the fact that SHA pinning guarantees code immutability but not its authorship or origin within the upstream project. The issue highlights a gap in current supply chain security practices for GitHub Actions.
iPhone Flaw Allowed Apps to Record Audio Despite Denied Microphone Permission
A patched flaw in Apple's audio service allowed iOS and iPadOS apps to record audio even after users denied microphone permission. The vulnerability, discovered by two security researchers, was fixed in security updates released on August 17, with no reported exploitation in the wild.
Six Unpatched Flaws in Danfoss Refrigeration Controllers Allow Remote Takeover
Six unpatched vulnerabilities in Danfoss AK-SM800A refrigeration controllers allow remote takeover without login, according to Dutch security consultancy Midnight Blue. The newest firmware version 5.4.0 does not fix these flaws, prompting cybersecurity agencies to advise removing management interfaces from the internet. These controllers manage store fridges and freezers and have been targeted in recent attacks causing defrost cycles and stock destruction.
CoffeeLoader Malware Spreads via Resurfaced Retadup Worm Variant
Cybersecurity researchers have identified a new variant of CoffeeLoader malware, which uses a patched 7-Zip library and a new version of the Armoury packer to deliver additional payloads. This variant establishes persistence and injects itself into legitimate processes to download and execute its main payload from a command-and-control server.
SANSAR Linux Cryptomining Malware Hides GPU Activity Using Modified NVIDIA Library
A financially motivated threat group, REF3364, used SANSAR Linux cryptomining malware to hide GPU activity by modifying NVIDIA Management Library (NVML) readings. This operation involved a Rust-based orchestrator, a kernel rootkit, and a backdoor that retrieves C2 addresses from an Ethereum smart contract. The malware falsifies GPU utilization to 0% and randomizes temperature and power readings to evade detection.
AI Redefines Security Operations, Shifting Analyst Roles to 'Threat Operator'
AI is increasingly automating tasks traditionally performed by security analysts, such as alert investigation and response actions. This shift will transform Security Operations Centers (SOCs) by integrating security functions and creating a new role called the 'Threat Operator'. Threat Operators will direct AI-driven operations, translate business risks into defensive priorities, and oversee AI system decisions across the security lifecycle.
Infostealer Logs Pose Evolving Operational Security Challenge for Organizations
Infostealer logs, containing compromised corporate credentials and session cookies, are increasingly challenging organizational security teams. These logs often originate from unmanaged personal devices, complicating incident response beyond simple password resets. The growing volume of these exposures necessitates advanced strategies for identifying active threats amidst numerous stale credentials.
Mandiant Details BREEZE COMET Threat Actor Targeting Brazilian Financial Services
Mandiant, part of Google Threat Intelligence Group, has identified BREEZE COMET (formerly UNC5669) as a financially motivated threat actor actively targeting Brazilian financial services, retail, and eCommerce organizations since early 2024. This group specializes in manipulating payment systems and banking software to conduct fraudulent transfers and is noted for using generative AI in malware development, potentially increasing the scale and sophistication of future operations.
Fake Remote Workers Exploit Hiring Processes to Infiltrate Corporate Networks
Security teams face a growing threat from fraudulent remote workers who exploit hiring processes to gain legitimate access to corporate networks. These individuals, sometimes linked to state-sponsored groups like those from North Korea, use various tactics to impersonate legitimate hires and exfiltrate sensitive data or conduct cybercriminal activities. This issue highlights a gap in identity verification during remote hiring, where traditional checks do not confirm the actual user of an account or device.
BO Team Exploits Log4Shell in VMware vCenter, Compromises Russian Firm's Code-Signing Server
The BO Team exploited a Log4Shell vulnerability in VMware vCenter to compromise a Russian software developer's network, including its code-signing server. This months-long intrusion led to the theft of confidential information and established persistent access, indicating a potential supply chain attack.
New Mior and Skyfall Attacks Bypass AMD SEV-SNP VM Protections
Researchers have disclosed two new attacks, Mior and Skyfall, that can compromise AMD SEV-SNP protected virtual machines. These attacks allow a malicious host to read data from within the VMs, with Mior also capable of modifying memory and faking attestation reports. AMD has released firmware and microcode updates to address these vulnerabilities.
Google Cloud warns of three AMD Turin CPU flaws affecting VMs, requiring guest OS patches
Google Cloud has identified three Transient Scheduler Attack (TSA) flaws in AMD Turin CPUs used in its VMs, which could allow data leakage between processes within a VM. Customers need to apply guest OS patches and perform a stop-and-start operation on affected VMs to enable mitigations. Google Cloud has already implemented mitigations against attacks between VMs and between VMs and the host.
Git Patches Code Execution Flaw in Recursive Clones (CVE-2026-111760)
Git has released patches for a flaw, CVE-2026-111760, that allowed malicious repositories to execute code during recursive clones by manipulating `.git` files in submodules. The fix prevents extremely long paths from being truncated, which previously enabled attackers to replace `.git` files with symbolic links to malicious hooks. Users are advised to update their Git installations and related tools to the patched versions.
Default 'admin:admin' Password in ForgeRock Servers Allowed Car App Hacks
Security researchers exploited a default 'admin:admin' password on ForgeRock identity platform servers used by two car brands, gaining full access to customer accounts. This vulnerability allowed them to unlock and start connected vehicles, demonstrating a critical flaw in deployment security.
Web3Auth Flaw Leads to Keplr Crypto Theft; Provider Empties At-Risk Wallets
A security flaw in Web3Auth, used by Keplr for Google and Apple sign-ins, allowed an attacker to steal cryptocurrency from some Keplr wallets. Web3Auth subsequently moved funds from other vulnerable wallets to recovery addresses to prevent further theft. Users who created Keplr wallets with Google or Apple accounts are advised to move unstaked funds to more secure wallet types.
LNbits Urges Operators to Shut Down Servers Due to Active Exploits
LNbits, an open-source Bitcoin Lightning wallet system, has advised all operators to shut down their servers following a wave of attacks. The project is preparing security fixes, with one operator reporting a loss of 0.023 bitcoin from an exploit before the shutdown notice was issued.
Athena AI to Begin Disclosing 50 Vulnerabilities This Month, Joins Anthropic's Project Glasswing
Athena, an AI system, will start disclosing its first 50 vulnerability findings this month, having identified over 40,000 findings across 500+ projects since June. The project has also joined Anthropic's Project Glasswing to further its vulnerability discovery efforts and plans to open-source its infrastructure for disclosure and patching. This initiative addresses the challenge of responsibly managing and disclosing AI-discovered vulnerabilities at scale.
Backdoor Found in Control Web Panel Server Allows Undetected Command Execution
Security researchers discovered a backdoor, named cwpdoor, hidden within the web server component of Control Web Panel (CWP), a Linux hosting management tool. This backdoor allows attackers to execute commands via web requests without leaving traces in access logs, indicating active exploitation in the wild.
APT Flaw in Debian and Ubuntu Allows MitM Attackers to Swap Verified Packages
A flaw in APT, the package manager for Debian and Ubuntu, allows attackers to swap verified package downloads with malicious code, which then executes as root. This vulnerability, CVE-2026-109518, affects Debian 12 and Ubuntu 22.04 and later by default, and requires APT to retry failed downloads. Debian has released fixed versions, and Ubuntu is distributing updates for affected releases.
Critical Ruby on Rails Flaw Allows Arbitrary File Write via Crafted ImageMagick Uploads
A critical vulnerability in Ruby on Rails' Active Storage component allows unauthenticated attackers to write arbitrary files to a server by uploading a crafted file. This flaw, rated 9.8 on the CVSS scale, affects applications that use ImageMagick for image processing and could lead to remote code execution or lateral movement by exposing sensitive application credentials.
Google Patches Critical Android RCE Flaw Affecting Android 14-17
Google released patches for a critical Android vulnerability, CVE-2026-58868, that could allow remote code execution without user interaction. The flaw affects Android versions 14 through 17 and is addressed in security patch level 2026-11-01 or later, which device manufacturers distribute.
DarkBlinders Compromises Israeli Individual and Iraqi Kurdish Government Cloud
An Iran-aligned threat actor, DarkBlinders, compromised a high-profile individual in Israel and a government cloud environment in Iraqi Kurdistan between August and October 2026. The group uses spear-phishing to deliver trojanized utilities or direct downloads, leading to the deployment of SHELBYLOADER and SHELBYC2 backdoors.
Telegram Android Flaw Allowed Remote Code Execution via Voice Calls
A flaw in Telegram for Android, present in versions 8.6.2 through 12.10.3, allowed remote code execution when users answered voice calls. This vulnerability, if chained with other bugs, could lead to full phone control, and was fixed in version 12.10.4.
26,400 Valid Cloud and AI Credentials Found Exposed in Public Gitee Repositories
A researcher discovered 26,400 active cloud and AI service credentials exposed in public Gitee repositories, which were still functional when tested. These credentials, including keys for Alibaba Cloud, YouTube API, DeepSeek, Zhipu, and OpenAI, could allow unauthorized access and potential misuse of services. Developers who have committed keys to public Gitee repositories should consider them compromised.
Critical 16-Year-Old UEFI Flaw Allows Code Execution Before Secure Boot on Network-Booting Servers
A 16-year-old critical flaw in EDK II, the open-source code for UEFI firmware, allows an attacker to run code before the operating system loads on machines booting over a network. The vulnerability, CVE-2026-97631, is a buffer overflow in the network-boot code (PXE) and has a severity rating of 9.5 out of 10.
Microsoft Authenticator Android Flaw Allowed Unauthorized Token Theft by Malicious Apps
Microsoft fixed a vulnerability (CVE-2026-80097) in its Authenticator app for Android that allowed unprivileged malicious applications to impersonate legitimate Microsoft apps and steal work account sign-in tokens. The flaw, rated 8.6 CVSS, could grant full access to email, files, and calendars, but required user interaction and has not been exploited in the wild.