GoldDigger is a sophisticated Android banking trojan that primarily targets mobile banking users in South Africa and across Europe, with evidence suggesting plans for global expansion. The malware employs advanced evasion techniques including a custom packer called 'dpt-shell', anti-debugging mechanisms, and Frida detection. It disguises itself as legitimate airline and shopping applications to deceive victims. GoldDigger exploits Android Accessibility services to perform on-device fraud, steal credentials, intercept SMS-based two-factor authentication, and execute unauthorized transactions. A unique feature is its ability to run targeted banking applications in a virtual environment, allowing complete interception of API calls and runtime behavior. The malware maintains communication with command-and-control servers via encrypted WebSocket protocol, enabling capabilities including screen recording, audio capture, phishing overlays, and remote device manipulation.
A new NFC relay malware family called WindRelay has been discovered operating in combination with SpyNote RAT to enable sophisticated contactless payment fraud. The scheme uses live social engineering phone calls where fraudsters impersonate bank employees and guide victims to install personalized RAT malware labeled with the victim's own name. Once installed, the RAT enables silent deployment of WindRelay, which captures contactless payment card data via NFC when victims tap their cards to their phones. The captured data is relayed in real-time to fraudster-controlled terminals for immediate cash-out through physical purchases or ATM withdrawals. The operation employs dual monetization, combining RAT-driven digital loan fraud with NFC-based card-present transactions. Group-IB identified 23 WindRelay samples targeting victims in Czechia, Slovakia, and Slovenia between November 2025 and July 2026.
A campaign uses fake websites impersonating CNN, Stremio, and Avast to distribute legitimate remote administration software O&O Syspectr pre-linked to attacker accounts. The lookalike sites closely mimic authentic homepages and trick Windows users into downloading installers that appear legitimate but grant attackers remote access to victim computers. Additional fake sites use cryptocurrency mining game lures to distribute the same tool. All installers are digitally signed legitimate software, making antivirus detection difficult. The campaign uses multiple Syspectr account IDs embedded in filenames, with CNN, Avast, and Stremio lures sharing one account while crypto-mining lures use another. O&O Software responded by disabling Remote Desktop and Remote Console access for free accounts and suspending the abusive accounts.
On August 4, 2026, ChainDrop, a self-propagating worm variant of Mini Shai-Hulud linked to TeamPCP, infiltrated the npm ecosystem through a compromised maintainer account of the keyv ecosystem. The attacker injected malicious code into GitHub repositories, weaponizing legitimate CI/CD pipelines to publish poisoned packages with valid SLSA Build Level 3 provenance attestations, making them indistinguishable from clean releases. ChainDrop spread to over 400 packages within four hours by stealing npm tokens and republishing infected versions. The worm employs Ethereum smart contracts for C2 infrastructure, enabling domain rotation without modifying deployed malware. It features destructive capabilities, wiping victim home directories upon token revocation, and achieves persistence through IDE and AI-agent configuration files. The payload harvests credentials from npm, GitHub, AWS, Azure, GCP, Kubernetes, HashiCorp Vault, and other services, exfiltrating data via GitHub repositories and EtherHiding techniques.
A fraudulent version of the widely-used PC cleaning utility CCleaner is being distributed through a convincing imitation website to deploy GhostDesk, a malicious Chrome extension functioning as spyware. The attack begins when users download the fake application from a lookalike site, which then launches a multi-stage infection using CScript to modify Chrome's Security Extension and install malicious components. Once active, GhostDesk performs extensive surveillance including credential theft, keylogging, screenshot capture, cookie harvesting, and cryptojacking. The extension establishes command-and-control communications via WebSocket connections and can execute arbitrary code within browser tabs. Similar fake versions of other popular software like 7-Zip and Adobe Acrobat have been identified using identical infection techniques.
An investigation uncovered a sophisticated infection chain beginning with a ClickFix lure and utilizing a legitimately signed IBM SPSS IDE alongside four decoy DLLs and a date-formatting API as a trampoline. This chain deploys BabaDeda loader stage that ultimately delivers CNCMachineRMS, a 1.14 MB x64 remote administration implant with no imports and runtime-built strings. The implant provides operators with comprehensive remote access capabilities including an interactive shell, file manager, screen capture, local account backdoor, and seven persistence mechanisms. It employs a custom scripting language and uses the same binary container format for configuration and C2 traffic. The implant beacons every 600 seconds, creates privileged local accounts, and supports twenty typed commands for downloading and executing additional payloads, indicating hands-on-keyboard access with follow-on stages determining actual damage.
A modular espionage framework targeting entities in Israel has evolved to incorporate sophisticated command-and-control capabilities. The framework employs DNS A-record responses to dynamically select between direct HTTPS connections and a Google Apps Script relay for each transaction, enabling operators to rotate communication channels and deployment identifiers. The communication module uses DNS infrastructure to validate and update Google Apps Script deployment IDs, while XOR encoding obfuscates command-and-control traffic. An inter-component broker coordinates framework DLL components, enabling runtime upgrades without system restarts. The infrastructure leveraged a previously expired Israeli domain, now repurposed with custom authoritative DNS servers, alongside legitimate Google services to blend malicious traffic with normal network activity.
The Head Mare APT group exploited a chain of vulnerabilities in TrueConf video conferencing servers to deploy PhantomCore and PhantomGraph backdoors. Attackers connected to unpatched TrueConf servers via port 4307/TCP without authorization, using vulnerabilities KLCERT-26-057 and KLCERT-26-058 to execute arbitrary code with NT AUTHORITY\SYSTEM privileges. They replaced legitimate TrueConf client installers with infected versions containing PhantomCore, and deployed a web shell for persistent access. The PhantomGraph backdoor utilized Microsoft OneDrive as command-and-control infrastructure. Affected TrueConf versions included 5.3.X through 5.3.9, 5.4.X through 5.4.9, and 5.5.X through 5.5.5. Multiple Russian organizations across various industries were targeted, including instrument manufacturing, electronics, transportation, energy, IT, and software development. The vulnerabilities were patched in June 2026.
A large-scale software supply chain attack compromised over 400 npm packages through a self-propagating worm called ChainDrop, a new variant of Mini Shai-Hulud. The campaign exploits stolen npm publishing credentials to automatically modify and republish legitimate software releases. ChainDrop targets developer workstations and CI/CD environments, harvesting credentials from npm, GitHub, AWS, Kubernetes, and HashiCorp Vault before validating access and enumerating resources. The malware uses preinstall lifecycle scripts for automatic execution, establishes persistence through repository configuration modifications, and abuses GitHub Actions OIDC trusted publishing workflows. After stealing credentials, it autonomously propagates by downloading packages, inserting malicious payloads, and republishing them with incremented versions, demonstrating how compromised developer identities can enable widespread ecosystem compromise.
Aeternum is a C++ botnet loader utilizing the Polygon blockchain for command-and-control infrastructure instead of traditional centralized servers. Threat actors write encrypted and plaintext instructions directly to smart contracts, which infected devices query via public RPC endpoints. The malware implements weak PBKDF2HMAC/AES-GCM encryption with self-salting passwords, allowing payload decryption using only the smart contract address. Analysis reveals three related samples: the core Aeternum loader with Telegram-based exfiltration, a blended threat combining XWorm RAT with XMRig cryptocurrency miner, and Python source code revealing anti-analysis checks and cryptocurrency wallet targeting. The botnet demonstrates resilience through decentralized infrastructure, making traditional law enforcement takedowns significantly more challenging while maintaining low operational costs for attackers.
In late June 2026, a new malware family named Abyssos was identified, representing a modular remote administration tool written in C++ with diverse capabilities including credential theft, file exfiltration, and remote access via VNC. The malware employs LLVM-based obfuscation techniques such as control flow flattening and string encryption to evade security products and complicate analysis. Abyssos uses a custom TCP protocol with AES-GCM encryption for network communication and supports numerous commands for system manipulation, data collection, and module deployment. It features anti-analysis mechanisms detecting hypervisors and security tools, though recent versions lack these checks. The malware demonstrates active development with multiple versions implementing different obfuscation passes, suggesting continued evolution of its capabilities and evasion techniques.
A threat hunting investigation identified suspicious PowerShell content served from an IP address (203.188.171.166) and domain (dorenzaa.com), both retrieving ZIP archives from Vercel-hosted infrastructure. The PowerShell loaders extract and execute payloads locally, including Grape.exe, UltraToolliteSetup.exe, and draw.io.exe. Analysis revealed heavily obfuscated PowerShell stages utilizing Base64 encoding, XOR-based obfuscation with the key 'Write', dynamically constructed IEX commands, and hidden PowerShell execution. A decoy 'Verification complete!' message disguised as Google.com was presented to victims during execution. Multiple Vercel instances hosted additional artifacts including loader scripts and executables. The initial infection vector remains unidentified, suggesting these PowerShell-hosting URLs represent second-stage delivery points in a multi-stage attack chain.
The Kimsuky threat group has integrated artificial intelligence capabilities into its attack operations, establishing local large language model environments using Ollama, GPT4All, and Msty. Evidence indicates the group is accumulating technologies to incorporate AI across attack operations, including AI-generated decoy documents and retrieval-augmented generation for document analysis. The campaign, dubbed Operation GitPower, continues targeting foreign diplomatic missions and sectors including military, security, and virtual assets. Attacks utilize malicious LNK files contained in ZIP archives, executing obfuscated PowerShell scripts that abuse Git-based repositories as command-and-control infrastructure. The group distributes encrypted AsyncRAT payloads disguised as image files through GitHub. Linguistic indicators including North Korean vocabulary patterns such as "싸이트", "가입리력", and "로출되였는지" support attribution to North Korean state-sponsored operations under the Reconnaissance General Bureau.
A self-propagating npm worm dubbed ChainDrop infected over 400 packages downloaded hundreds of millions of times weekly, including popular packages like keyv and cacheable-request. The worm steals cloud credentials, npm and GitHub tokens, SSH keys, and sensitive developer data while extracting temporary credentials from GitHub Actions runner memory. It uses stolen npm publishing tokens to infect additional packages while maintaining their legitimate functionality. The attackers established persistence through VS Code and Claude Code configurations, employed blockchain-based command-and-control resolution via Ethereum smart contracts, and can execute attacker-supplied code. The operator demonstrated ability to silently reconfigure C2 infrastructure through Ethereum transactions without updating deployed instances. ChainDrop employs three layers of obfuscation and encryption, exfiltrates data through encrypted channels, and publishes stolen tokens in public commit messages.
Security researchers have uncovered a sophisticated cyber espionage operation deploying two previously undocumented malware families, OctLurk and SilkLurk, targeting government and public-sector organizations across Central Asia and the Middle East. Both modular backdoors utilize victim-specific decryption mechanisms, extensive obfuscation, and in-memory execution to evade detection. The malware enables credential theft, remote access, network reconnaissance, and plugin-based expansion. Operations began in January 2025, affecting entities in Afghanistan, Kazakhstan, Kyrgyzstan, Syria, Tajikistan, and Uzbekistan. Victims include government offices, foreign affairs ministries, law enforcement agencies, healthcare providers, logistics organizations, research institutions, urban planning facilities, and educational establishments. Attackers deployed additional tools including Impacket's SecretsDump, Browser Password Decryptor, Pandora RC, Fscan, WinRAR, 7-Zip, and PlugX. A companion utility, LurkProxy, proxies a
A sophisticated macOS campaign has been discovered using a fake Zoom installer to deploy Overlord RAT, an open-source remote access framework. The attack employs a .NET-based downloader disguised as ZoomMeetings, representing an uncommon approach for macOS threats. The multi-stage attack fingerprints the victim's system to deliver platform-specific payloads for macOS ARM64, macOS Intel, or Windows from attacker-controlled infrastructure. The second stage deploys Overlord RAT with extensive capabilities including keylogging, screen capture, audio and webcam access, filesystem manipulation, and remote desktop streaming. The malware communicates with command-and-control servers over encrypted WebSockets and maintains persistence through LaunchAgents. The campaign shares characteristics with previous North Korean operations, including similarities to FlexibleFerret malware and the Contagious Interview campaign.
A sophisticated macOS malware campaign leverages ClickFix social engineering to infect victims. The attack begins with a fake CAPTCHA prompt delivered via email links, tricking users into executing malicious commands in Terminal. This downloads a profiling script that collects system information and deploys architecture-specific Go-based Mach-O payloads. The stealer targets browser passwords, Apple Keychain credentials, and cryptocurrency wallets. Its most notable feature is a DRAIN function that gradually siphons cryptocurrency from victims' wallets by redirecting portions to attacker-controlled accounts. The malware supports Bitcoin, Litecoin, Dogecoin, Monero, Ethereum, and XRP. Infrastructure analysis reveals hosting through Aeza Group, a sanctioned Russian bulletproof hosting provider. The malware achieves persistence through macOS Background Task Management and uses various evasion techniques including Gatekeeper bypass and credential harvesting via fake system prompts.
Vanta Stealer is a Python-based information stealer utilizing PyArmor protection and PyInstaller packaging to complicate defensive analysis. The malware systematically harvests credentials from Chromium-based browsers, communication platforms like Discord and Telegram, gaming applications including Steam, Riot Games, Roblox and Minecraft, cryptocurrency wallets, Mullvad VPN configurations, and sensitive documents. It performs token enrichment by validating stolen Discord credentials against the API to retrieve account details, billing information, Nitro status, and server privileges. The modular architecture downloads dedicated browser extraction utilities at runtime, maintains independence between collection modules, and generates structured inventory reports before consolidating harvested data into ZIP archives. Exfiltration occurs via HTTP POST to attacker-controlled infrastructure with victim metadata. Distribution likely occurs through social engineering campaigns involving phishing emails, trojanized...
A sophisticated software supply chain attack named ChainDrop has infected hundreds of npm packages, including popular caching libraries with millions of weekly downloads. Beginning August 4, 2026, attackers compromised a GitHub account of a keyv package maintainer, injecting malicious code into legitimate repositories. The malware executes credential-stealing payloads targeting developer workstations and CI/CD runners, harvesting npm tokens, GitHub credentials, cloud access keys, SSH keys, and database credentials. Using stolen credentials, the worm self-propagates by compromising additional repositories and publishing poisoned packages with valid provenance attestations. ChainDrop employs Bun runtime for execution, establishes persistence through developer tool configurations, and exfiltrates encrypted data using blockchain-based command-and-control infrastructure. This campaign represents an evolution of the Shai-Hulud npm worm.
Cybercriminals are exploiting API keys used by developers to access AI platforms through a technique called token jacking. Attackers steal these authentication tokens to gain unauthorized access to expensive AI resources, which they either use themselves or resell through gray-market services called transfer stations. These transfer stations act as intermediaries, offering frontier AI model access at discounted rates using stolen credentials. The financial impact can be catastrophic, with victims potentially losing hundreds of thousands to millions of dollars before detection due to unlimited scaling defaults and cyclical billing. Attackers obtain tokens through information stealers, phishing campaigns, compromised code repositories, and poisoned npm packages. Organizations can mitigate risks through spending limits, privileged account reviews, short-term bearer tokens, AI gateways, and tight development environment management.
