From Portuguese sources
Translated from Portuguese
Atlassian Fixes Critical Path Traversal Flaw CVE-2026-21589 Exposing Files in Jira and Confluence
Vulnerabilities & Exploits
Apache Struts CVE-2026-104711 Enables Remote Code Execution via Legacy RESTful Mapper
Vulnerabilities & ExploitsWeb Application Vulnerabilities Surge as AI-Driven Development Outpaces Security Testing
Vulnerabilities & ExploitsMicrosoft Fixes CVE-2026-96940 in Exchange Server Allowing Authenticated Mailbox Access
Microsoft has patched CVE-2026-96940, a CVSS 8.8 vulnerability in Exchange Server that lets any authenticated user read other users' mailboxes without administrative rights. The flaw exposes full message content and attachments including contracts, spreadsheets, and sensitive documents. Affected on-premises versions include Exchange Server Subscription Edition RTM, Exchange 2016 CU23, Exchange 2019 CU15, and Exchange 2019 CU14. Exchange Online users are protected because the fix was applied server-side. Microsoft rates exploitation as likely but reports no confirmed attacks in the wild at disclosure time. The issue turns a single low-privilege credential into broad access to executive, legal, and financial correspondence.
Realtek Jungle SDK Flaw CVE-2021-35394 Fuels Cling Botnet Spread Across Routers
Researchers at Nozomi Networks have observed a sharp rise in exploitation attempts against CVE-2021-35394, a critical remote code execution vulnerability in the Realtek Jungle SDK. The flaw, rated 9.8 on the CVSS scale and disclosed five years ago, is being used to deploy the Cling botnet on routers and video recorders. The affected SDK is embedded in products from multiple vendors, leaving large numbers of devices exposed because firmware updates are rarely applied. Cling carries exploits for seven distinct vulnerabilities targeting Realtek, Linksys, MVPower, TBK, LB-LINK, FiberHome and China Mobile hardware. Once installed, the malware performs recursive scanning, spreads like a worm, manipulates TCP tunnels and proxies, and participates in DDoS attacks. Its command-and-control channel hides instructions inside STUN protocol transaction IDs, impersonating legitimate responses from Google public STUN servers. FortiGuard Labs has confirmed the findings and tracks the variant as ClingSTUN.
ThreatMon Exposes SQL Server xp_cmdshell Abuse Stealing Credentials from Viva Aerobus
Researchers at ThreatMon discovered an exposed attacker infrastructure containing 17 attack tools and data stolen from airline Viva Aerobus. Attackers gained operating system access through the xp_cmdshell feature in SQL Server, which allows execution of system commands when enabled. They issued Windows and encoded PowerShell commands directly through database sessions. Data exfiltration occurred by reading files, splitting content into smaller chunks, converting to Base64, and returning results via normal SQL query responses to avoid network detection. Recovered materials included browser, Windows, and SQL credentials, source code, configuration files, database connection strings, OAuth, email, SFTP, and payment system references, plus connection history from SQL Server Management Studio. Activity took place between September 25 and 29, with no identified initial access vector or link to known malware families.
Dell Patches Six Critical Flaws in Container Storage Modules for Kubernetes
Dell has fixed six vulnerabilities in its Container Storage Modules that integrate storage systems with Kubernetes clusters. Two of the issues received the maximum CVSS score of 10.0, allowing remote unauthenticated attackers to obtain full administrative credentials for registered storage backends. Additional flaws enable privilege escalation to root on cluster nodes, exposure of hardcoded credentials, and leakage of Kubernetes secrets across the entire cluster. All versions prior to 1.17.0 are affected, with the fixes delivered in version 1.18.0. No workarounds exist, and Dell recommends rotating JWT signing keys after applying the update because the previous keys must be considered compromised.
International Law Enforcement Operation Dismantles KillSec Ransomware Group and Seizes 110 TB of Stolen Data
An international operation coordinated by Eurojust with support from Europol has dismantled the infrastructure of the KillSec ransomware group. Authorities seized five servers containing at least 110 terabytes of data stolen from victims and took control of the group's leak site domains. Three individuals were arrested, including a 16-year-old identified as the group's primary administrator and operator. The coordinated action involved law enforcement from nine countries and included eight searches across Spain, Greece, the United Kingdom, and Romania. KillSec has been active since 2024 and is linked to nearly one thousand ransomware incidents worldwide, primarily using a double-extortion model that combines data encryption with threats to publish stolen information. The seized data volume highlights the scale of the group's operations, which frequently targeted healthcare environments where system downtime directly impacts patient care.
Attackers Abuse Legitimate Microsoft Defender Exclusions to Conceal Malware
Huntress researchers have detailed an evasion technique in which threat actors avoid disabling Microsoft Defender entirely. Instead, they create targeted exclusions for specific folders or file extensions, allowing malware to operate undetected while the protection status remains apparently active. These exclusions are configured through PowerShell commands, Windows Management Instrumentation, Group Policy, or direct registry modifications, all requiring administrator privileges after initial compromise. A registry key named HideExclusionsFromLocalAdmins can further conceal the list of exclusions from local administrators viewing the interface. The approach has been linked to campaigns involving GootKit in 2019, WhisperGate in 2022 that excluded the entire C: drive, and Muddled Libra in 2024. Defenders are advised to monitor registry changes directly, as this bypasses interface hiding, and to flag exclusions of entire drives or common directories such as temporary and downloads folders.
Cisco Confirms Active Exploitation of CVE-2026-76504 in Catalyst SD-WAN Manager
Cisco has confirmed that the critical vulnerability CVE-2026-76504 in Catalyst SD-WAN Manager has been exploited in attacks throughout September. The flaw carries a CVSS score of 9.8 and allows attackers to gain full administrator access without any credentials by bypassing API authentication through malformed URI encoding. The issue affects the login session handling mechanism, enabling forged requests to grant netadmin privileges by default. Similar URI manipulation techniques were observed earlier this month in Oracle PeopleSoft attacks. Patches are available across multiple release trains including 20.9.10.1, 20.12.8.2, 20.15.6.1, 20.18.4.1, 26.1.2.1 and 26.2.1, while older installations must migrate to supported versions. Cisco recommends restricting access to trusted hosts and placing the Manager behind firewalls until updates are applied. The vulnerability was discovered during routine support operations and verified by Cisco's Product Security Incident Response Team.
Zero-Days in PaperCut MF Chained to Compromise Active Directory in Education Sector
Analysts at eSentire investigated an attack that chained two zero-days in PaperCut MF, tracked as CVE-2026-81578 and CVE-2026-82078, to move from an internet-exposed print server to a domain controller in an education sector client in under two days. Attackers gained initial access through the card or badge query field, delivering malicious Java code that allowed unauthenticated configuration changes and arbitrary bytecode execution on version 24.0.2. The first stage loader reassembled payload fragments in memory, launched the next stage, and deleted its own files while remaining compatible with multiple Tomcat versions. A web shell followed, accepting commands via a custom HTTP header, reading configurations, and erasing log traces while positioning itself early in the request processing chain. The command-and-control implant was AdaptixC2, hidden inside a modified Microsoft Copilot binary downloaded from Alibaba Cloud infrastructure. Privilege escalation was achieved without passwords by locating a domain-privileged service account, stealing its access token, and relaunching the implant under those rights. On the domain controller the payload arrived via administrative share and was executed by modifying the Windows PlugPlay service, after which the legitimate path was restored to minimize traces. The operators extracted credentials from memory and registry, enabled Restricted Admin mode, used an NTLM hash for RDP access, and copied the full Active Directory database containing passwords for all domain accounts.
WatchGuard Patches Critical API Flaws in Access Points Allowing Unauthenticated Command Execution
WatchGuard has released firmware version 3.4.8 to address three vulnerabilities affecting its access points, two of which are rated critical. The most severe issues, CVE-2026-86102 and CVE-2026-101891, both scored 9.3 on the CVSS scale and reside in the internal API management service. These flaws permit unauthenticated attackers with network access to execute arbitrary shell commands or bypass access controls without requiring any credentials or valid sessions. A third vulnerability, CVE-2026-87969 with a CVSS score of 8.6, involves command injection through the command-line diagnostic interface but requires authenticated administrator privileges. All firmware versions from 1.0 through 3.4.7 are impacted. The vulnerabilities were disclosed on September 28 with no evidence of active exploitation or public proof-of-concept code available at the time. A compromised access point could serve as an internal foothold for further network attacks due to its visibility into corporate traffic.
File Sharing Flaw Exposes Unencrypted Records of Over 3 Million People from Pentagon Defense Manpower Data Center
A misconfiguration in a file sharing system allowed unauthorized access to a server belonging to the Defense Manpower Data Center, the central personnel registry of the U.S. Department of Defense. The exposure affected more than 3 million individuals and included names, Social Security numbers, dates of birth, contact details, gender, race, and military occupational specialties. The data remained unencrypted throughout the incident window that lasted from October 2025 until July 2026. The DMDC discovered the issue on 16 July 2026, applied an immediate fix, and restored the system, yet victim notifications did not begin until 18 September. Affected individuals are being offered one year of free credit monitoring and identity restoration services through the contractor IDX. No evidence of data misuse has been identified and no attribution has been made public.
NVIDIA Unveils Open Agent Safety Platform to Secure Autonomous AI Agents
NVIDIA announced the Open Agent Safety Platform on September 28, introducing a set of tools designed to contain autonomous AI agents that interact with models, tools, code execution environments, data, networks, and corporate systems. The platform consists of two main components: the open-source OpenShell runtime under Apache 2.0 license, which isolates agents at the kernel level, and NVIDIA Sentry, which performs monitoring and policy enforcement inside BlueField data processing units. This hardware separation ensures that security controls remain effective even if the agent's host environment is compromised. The architecture is structured in three layers covering the application, runtime governance, and underlying infrastructure. Pre-execution verification combined with real-time behavioral monitoring restricts actions that deviate from defined policies. The BlueField-4 DPU sits between agents and reasoning models, while the solution is optimized for Vera processors and BlueField DPUs with declared compatibility for other hardware. More than 100 organizations have expressed support for the initiative, although no performance metrics or independent test results were provided.
CISA Adds Two Critical Citrix NetScaler Flaws to Known Exploited Vulnerabilities Catalog
The CISA has added two critical vulnerabilities affecting Citrix NetScaler ADC and Gateway to its Known Exploited Vulnerabilities catalog following reports of active global exploitation. Both CVE-2026-88771 and CVE-2026-88772 carry a CVSS score of 9.5 and were patched by Citrix on September 27, the same day they were added to the catalog. The first flaw stems from improper input validation and allows unauthenticated arbitrary command execution on default installations. The second issue involves a buffer overflow that can lead to remote code execution or denial of service when DTLS is enabled on VPN virtual servers. Affected versions include 14.1-73.32, 13.1-63.21 and earlier, with fixes available in 14.1-73.37, 13.1-64.23 and later releases including FIPS variants. The issues were identified by watchTowr on September 26, and Citrix confirmed ongoing attacks against unpatched systems. Organizations are advised to apply patches immediately while preserving evidence and following full incident response procedures.
CSRF Flaw in Elementor Allows Admin Account Creation on Over 2 Million WordPress Sites
A critical CSRF vulnerability has been identified in the Elementor page builder plugin for WordPress, affecting versions 4.3.0 and 4.3.1. The issue enables attackers to create administrator accounts on target sites by exploiting improper handling of CSRF protections in the Editor Events module. With a CVSS score of 8.8, the flaw impacts more than 2 million installations out of over 10 million total WordPress sites using the plugin. No CVE identifier had been assigned at the time of disclosure on September 26. The vulnerability was fixed in version 4.3.2, and users are urged to update immediately. The discovery is credited to researcher Saggre following responsible disclosure, with no public records of active exploitation noted at the time of analysis.
Mandiant Uncovers WAF Bypass Campaign Exploiting Critical Oracle PeopleSoft CVE-2026-35273
Mandiant has identified an active campaign abusing CVE-2026-35273, a critical unauthenticated remote code execution flaw in Oracle PeopleSoft's Environment Management Hub (PSEMHUB). Attackers bypass web application firewalls by replacing the literal path /PSEMHUB/ with /%50SEMHUB/, exploiting the fact that many WAF rules inspect the URL before decoding while the PeopleSoft application server decodes it afterward. The exploitation chain relies on Java object deserialization via POST requests to /%50SEMHUB/hub, allowing deployment of two distinct JSP web shells. The group then establishes persistence with a trojanized installer that drops the SIDEEYE backdoor along with Neo-reGeorg and MeshAgent. Activity attributed to UNC6240, linked to ShinyHunters, began as a zero-day against educational institutions in June 2026 and has since expanded to higher education, technology, healthcare, agriculture, transportation, and government sectors.
Fake Terraform Providers on HashiCorp Registry Distribute Go Malware to Developers
Cybersecurity researchers have identified Go-based malware distributed through two fake Terraform providers and two Go modules hosted on the official HashiCorp registry. The providers gocommunity-io/dockerd and kreuzwenker/docker, along with modules gocommunity.io/orderedbtree and gogets.dev/btreex, impersonate legitimate projects and represent the first documented case of malicious code being delivered via the HashiCorp registry. Attackers approach developers on LinkedIn, Facebook, and job forums using fake Web3 company profiles, then supply seemingly harmless repositories whose malicious behavior is triggered through npm or PyPI dependencies. Once executed, the malware collects hardware attributes, operating system data, hostname, and node availability before sending the information to attacker infrastructure. Command and control relies on a Slack channel polled every ten seconds and encrypted commands read from Sepolia testnet Ethereum smart contracts every three seconds, with each infected client using ephemeral key pairs for targeted delivery. The code matches the Graphalgo campaign previously documented by ReversingLabs and attributed to North Korean actors.
Group-IB Discovers RemControl Android Banking Trojan Using Dynamic Overlay Attacks to Steal Banking PINs
Group-IB has identified a new Android banking trojan named RemControl with samples dating back to July 2026. The malware targets more than 30 financial institutions across Europe, the Middle East, and Canada, with Italy and France as primary focus areas. RemControl employs an overlay technique that displays a full-screen phishing replica of the legitimate banking app when the victim launches the real application, capturing PINs, access codes, and card details before seamlessly returning control to the authentic app. Phishing screens are not bundled in the APK but are fetched dynamically from operator-controlled servers reachable via Telegram channels, allowing rapid target changes without reinfection. Distribution relies on fake Google Play listing pages that mimic a streaming application, with geo-targeted delivery for Italian visitors. The trojan requests VPN permissions to block Play Store traffic and bypass real-time security checks, followed by accessibility service access that enables screen reading, keystroke logging, remote taps, and unlock pattern reconstruction.
Check Point Patches Critical Zero-Day CVE-2026-93616 Enabling Unauthenticated Remote Code Execution on Management Servers
Check Point has addressed a critical zero-day vulnerability tracked as CVE-2026-93616 with a CVSS score of 9.8 that allowed remote unauthenticated attackers to upload and execute arbitrary scripts on management servers. The flaw combined directory traversal with insecure file handling, enabling attackers to escape the intended directory and place scripts in executable locations. Affected products include Security Management Server, Multi-Domain Security Management Server, Log Server, Multi-Domain Log Server, and SmartEvent. Targeted attacks were observed starting July 23 before public disclosure, though exploitation remained limited. Vulnerable versions span multiple R82, R81, and R80 releases, with fixes now available through specific hotfixes and Jumbo Takes. Administrators are advised to apply patches immediately, place management servers behind a Security Gateway, and restrict TCP port 19009 access. Detection guidance includes reviewing cpm.elg logs for unusually long usernames and ReflectionUtils errors containing directory traversal sequences.
ShinyHunters Claims Breach of FBI Recruitment Portal and Demands Eight-Figure Ransom
The hacker group ShinyHunters has publicly claimed responsibility for compromising the FBI's official recruitment website, FBIjobs.gov, asserting access to sensitive data belonging to nearly all FBI agents as well as job applicants. According to the group, the intrusion extended to multiple internal systems including criminal justice databases, human resources platforms, and Medlink. The attackers stated they exploited a zero-day vulnerability in Oracle PeopleSoft to achieve remote code execution and subsequently defaced the careers site with a fabricated seizure notice. The FBI has acknowledged awareness of unauthorized activity on the portal but has not confirmed any data theft or the scope of the intrusion. ShinyHunters is now demanding an eight-figure ransom payment, framing the amount as a minor fraction of its own resources and warning that time is limited. The operation appears to be retaliation for an FBI public statement issued in May regarding the group's prior activities. Independent verification of the claims remains unavailable, and experts note that extortion groups routinely exaggerate the value of stolen data to increase pressure on victims.
WordPress Patches Critical Core Vulnerability CVE-2026-87902 Affecting All Versions from 4.7.0
WordPress has released patches for a critical vulnerability in the platform's core code, tracked as CVE-2026-87902 with a CVSS score of 9.2. The flaw impacts every version from 4.7.0 through 7.1.1, meaning virtually all unpatched installations remain exposed. Because the issue resides in the base system rather than plugins or themes, even sites without any extensions installed are at risk. The vulnerability allows an unauthenticated attacker to load arbitrary PHP files from outside theme directories, which can lead to remote code execution on certain server configurations. The bug was privately reported in July by researcher Robert Ressl through HackerOne and disclosed publicly on 22 September, with no confirmed exploitation observed so far. Administrators are urged to apply the updates immediately and keep automatic updates enabled.
PAYLOAD Ransomware Seizes Active Directory GPO to Disrupt Entire Windows Domain Without Encryption
Researchers at Kaspersky have documented an attack by the PAYLOAD ransomware that paralyzes an entire Windows domain without encrypting a single file. Instead of encryption, the operators leverage native Windows policy mechanisms to enforce disruption across the environment. The core of the attack is a malicious Group Policy Object named PAYLOAD linked directly to the root of the Active Directory domain. This placement allows the policy to reach virtually every connected device, turning it into a corporate-wide disruption tool. Through the GPO, the group distributes ransom notes from SYSVOL, replaces wallpapers and lock screens with extortion images displaying the message Welcome to Payload, and disables local administrator accounts on affected machines. A second policy object named win Firewall Off disables the Windows firewall across the entire fleet, increasing exposure during the operation. Initial access occurred in April 2026 via a compromised legitimate domain account on a FortiGate SSL VPN, with possible entry vectors including phishing, password spraying, and credential stuffing. The victim is a manufacturing company in the Middle East. The extortion model combines data theft with operational shutdown, delivering effects similar to traditional ransomware but without any decryption key to negotiate.
Critical Microsoft SharePoint Vulnerability Allows Remote Code Execution via SafeControls Bypass
A vulnerability tracked as CVE-2026-65660 with a CVSS score of 8.8 affects Microsoft SharePoint Server 2016, SharePoint Server 2019, and Subscription Edition. The flaw permits an authenticated low-privileged attacker to achieve arbitrary code execution on on-premises servers. It stems from improper handling of quotes in attacker-controlled Register directives within the ToolPane component, bypassing the SafeControls mechanism. This allows registration of dangerous .NET classes followed by deserialization-based remote code execution. Microsoft released patches on August 11, 2026, and initially reported no public disclosure or exploitation. Publication of a detailed technical analysis has now increased the risk for unpatched installations.
HackerSec's Yaga AI Agent Delivers Autonomous Penetration Testing with 98% Effectiveness
HackerSec has developed an autonomous penetration testing platform powered by its Yaga AI agent, marking a shift from scripted automation to adaptive, decision-making offensive security operations. Unlike traditional tools that follow predefined checks, Yaga observes the target environment, interprets responses, forms hypotheses, and replans attack paths in real time while maintaining memory and state across findings. The system relies on a proprietary harness that coordinates advanced AI models, security tools, contextual data, and ongoing decision logic rather than depending on a large language model alone. Yaga already achieves more than 98 percent effectiveness and supports continuous testing that matches the rapid change cycles of modern applications. CEO Andrew Martinez noted that periodic manual pentests can no longer keep pace with AI-accelerated attacks, requiring an AI-native model capable of operating at the same speed as the systems it protects. The platform allows human specialists to monitor and validate results while the agent conducts the investigation independently.
North Korean WaterPlum Group Infects 30,000 Computers Across 100 Countries via Fake Coding Tests
The North Korean-linked group WaterPlum, also known as Contagious Interview, has infected at least 30,000 computers in more than 100 countries between December 2025 and July 2026. According to the Internet Crime Complaint Center (IC3), the operation moved at least $10.7 million in cryptocurrency to North Korea and compromised more than 7,000 wallets. Attackers pose as recruiters on social media and freelance platforms, luring software developers with nonexistent job offers that require downloading malicious files disguised as coding tests. The malware arsenal includes BeaverTail, InvisibleFerret, OtterCookie, OtterCandy, and StoatWaffle, with the latter hiding inside blockchain-themed projects and abusing Visual Studio Code configurations. Developers are specifically targeted because they often store production keys, wallets, and credentials on the same machines used for testing third-party code. Japanese authorities have dismantled laptop farms connected to the scheme that helped maintain the attackers' operational cover.
EtherHiding Campaign Hides Banking Trojan C2 in Polygon Smart Contracts
Security researchers have detailed the EtherHiding campaign, which conceals command-and-control infrastructure inside smart contracts on the Polygon blockchain. The final payload is a malicious browser extension that functions as a banking trojan, intercepting credentials and two-factor codes from approximately 479 financial and cryptocurrency websites. Instead of embedding fixed addresses in its code, the malware queries an encrypted C2 server address from the smart contract, allowing operators to change destinations through low-cost blockchain transactions that bypass domain blocking. Infection begins when victims visit one of 31 compromised legitimate sites that inject JavaScript displaying a fake CAPTCHA prompt. The prompt instructs users to press Windows+R and execute a PowerShell command, a social engineering technique known as ClickFix that downloads the malicious scripts without exploiting any software vulnerability. The campaign has remained active from November 2025 through at least September 2026, demonstrating the resilience of blockchain-based infrastructure against traditional takedown methods.
Click2Shell Flaw in WordPress Core Enables Remote Code Execution via Single Malicious Link
Researchers at pwn.ai have disclosed Click2Shell, a vulnerability in the WordPress core that allows an attacker to install a malicious theme and achieve remote code execution simply by tricking an authenticated administrator into opening a crafted link. The isolated flaw carries a CVSS score of 7.1, but the full attack chain reaches 9.6. The issue stems from an interpretation mismatch between the WordPress.org theme directory and the administrator browser, causing the browser to automatically trigger the install button without any user confirmation or password prompt. Affected versions start from 6.0 and run up to but not including 7.1.1. The vulnerability has been fixed in WordPress 7.1.1 with backported patches released for all supported branches down to version 4.7. No exploitation in the wild had been observed at the time of disclosure, yet the low barrier of convincing an admin to click a link makes prompt patching essential.
SETTRA Ransomware Deploys MeshAgent and gdrv.sys BYOVD Against Windows Systems
Huntress analysts have identified two separate incidents involving the newly observed SETTRA ransomware targeting Windows environments. The attacks combined the legitimate MeshAgent remote management tool with a Bring Your Own Vulnerable Driver technique using gdrv.sys to disable security controls. Victims included a consumer services and retail company hit in July and an industrial organization compromised in September. Initial access occurred through VPN connections or previously stolen credentials, after which operators deployed MeshAgent to maintain persistence and execute commands. Before encryption, the group deleted Windows event logs, disabled system recovery features, removed recovery partitions, and overwrote free disk space. Encrypted files received the .locked or .locked_wip extensions, with the ransomware binary named after the victim domain and ransom notes dropped as RESTORE_FILES.txt.
HEAVYGRAM Spyware Uses Telegram Bots for Command and Control Against Iranian Targets
Researchers at Group-IB have published a detailed analysis of HEAVYGRAM, a spyware family that abuses the Telegram messaging platform as its command-and-control infrastructure. The malware family was first observed in the second half of 2023 and has since been linked with moderate confidence to the Handala Hack group. Instead of operating dedicated servers, the operators rely on Telegram bots, accounts, and groups to register infected hosts, receive commands, exfiltrate stolen data, and deliver additional payloads. Once active, HEAVYGRAM captures screenshots, records audio, harvests cached files, and steals data from Telegram Desktop installed on the victim machine. The campaign primarily targets Iranian journalists, dissidents, and individuals opposed to the Iranian government. Infection vectors include malicious files distributed via messengers, disguised as legitimate applications such as Pictory, KeePass, or Telegram-related tools, sometimes delivered as HTML applications or scripts.
BIND DNS Servers Receive Patches for 14 Vulnerabilities Including High-Severity DoS Flaws
The Internet Systems Consortium has released BIND 9 updates that address 14 vulnerabilities across multiple versions of the widely used DNS server software. Seven of the issues received a CVSS score of 7.5 and were rated high severity, while the remaining seven scored between 5.3 and 6.5. The most critical flaw allows an unauthenticated attacker to crash the named process with a single malformed SIG(0) request over DNS over HTTPS. Additional vulnerabilities enable cache poisoning through forged NXDOMAIN responses, downgraded secure delegations, and acceptance of unsigned answers, as well as resource exhaustion via uncontrolled cache growth and excessive CPU consumption. Affected releases span BIND 9.11.0 through 9.18.50, 9.20.0 through 9.20.27, and 9.21.0 through 9.21.25. No workarounds exist, making immediate upgrades to versions 9.20.29, 9.21.26, or 9.20.29-S1 the only mitigation. The ISC reports no known exploitation in the wild and none of the flaws appear in CISA’s Known Exploited Vulnerabilities catalog.
KREMLIN Banking Malware Hijacks Chrome and Edge to Steal Credentials and Session Tokens in Brazil
Elastic Security Labs researchers have detailed the operations of the KREMLIN banking malware, tracked under the identifier REF9334, which targets Chrome and Edge browsers to harvest credentials and session tokens. The campaign focuses almost exclusively on Brazil, with 98 percent of the 1,515 identified infections located in the country and impersonating a dozen Brazilian banks. Infection begins with multi-stage JavaScript loaders disguised as banking documents, invoices, or corporate papers that require manual execution by the victim. The loaders then deploy C++ installers and malicious browser extensions that modify the Secure Preferences file, enable developer mode, and overwrite protection objects with forged metadata using a technique called Phantom Extension. Once active, the extension collects session tokens, cookies, sessionStorage and localStorage data, 15 days of browsing history, screenshots, open tab information, and full HTML of visited pages. The operation has run since May 2025 across seven distinct campaigns and began using Ethereum smart contracts for infrastructure on 19 May 2026.
Cisco Confirms Active Exploitation of Critical Secure Email Gateway Flaw Allowing Root Command Execution
Cisco has confirmed active exploitation of a critical vulnerability in its Secure Email Gateway product that allows attackers to execute arbitrary commands with root privileges. The flaw, tracked as CVE-2026-76461 with a CVSS score of 9.8, stems from insufficient input validation in the message analysis logic of AsyncOS. Attackers can trigger the issue by sending a specially crafted email containing malicious SQL instructions, which leads to arbitrary SQL command execution and full root access on affected appliances. The vulnerability impacts physical and virtual deployments running AsyncOS versions 15.5 and earlier, 16.0, and 16.5. Patches are available in versions 15.5.5-0141, 16.0.4-302, and 16.5.0-780, while the Secure Email and Web Manager and Secure Web Appliance remain unaffected. CISA added the CVE to its Known Exploited Vulnerabilities catalog on September 14 and set a September 17 remediation deadline for federal agencies.
AWS Systems Manager Agent Flaw Allows Bypass of Port Forwarding Restrictions
Amazon has fixed a vulnerability in the AWS Systems Manager agent that permits an authenticated attacker to bypass restrictions on port forwarding sessions. The issue, tracked as CVE-2026-89049, stems from inadequate validation of equivalent representations of blocked link-local addresses. As a result, attackers can reach the EC2 instance metadata service at 169.254.169.254 and obtain temporary credentials tied to the instance's IAM role. Affected versions are those prior to 3.3.4851.0, with the fix included in that release and all subsequent versions. AWS recommends reviewing custom builds of the agent, restricting access to the AWS-StartPortForwardingSessionToRemoteHost document, auditing Session Manager activity, and applying least-privilege principles to IAM roles.
Malicious Twitch Extension Steals OAuth Tokens from Nearly 31,000 Users
A browser extension posing as an enhancement for Twitch has been stealing OAuth authentication tokens from approximately 31,000 users across Chrome and Firefox. The extension, known as Twitch Enhanced Viewer or JeetBot, was discovered by researcher Kush Pandya of Socket. It promised 1080p streaming in restricted regions and an ad-free experience while covertly exfiltrating session tokens to attacker-controlled servers. The tokens were transmitted in plaintext via network-layer redirects, allowing full access to chat functions, private messages, and account settings. Both the Chrome version with around 30,000 installations since June 26, 2025, and the Firefox version with 604 users since July 7, 2025, remained available in official stores at the time of reporting. Users are advised to immediately disable the extension and revoke active Twitch sessions to invalidate stolen tokens.
Revolut Confirms Limited Data Exposure After Fraudulent Government Email Request
Revolut has confirmed that a small number of customers had personal documents and full transaction histories exposed after the company responded to a fraudulent request that appeared to come from a government authority. The incident, disclosed on September 13, involved no intrusion into Revolut systems. Attackers instead used an unauthorized email account that carried valid credentials from a government domain, making the request appear legitimate. Exposed data included passport and driver’s license copies, identity verification selfies, full names, dates of birth, occupations, addresses, contact details, IBAN statements, complete transaction histories, and Bitcoin activity records. The combination of identity documents and detailed financial history significantly raises the risk of account takeover and targeted social engineering. Revolut stated that customer funds remained secure, blocked the malicious email source, notified regulators, and contacted all affected users.
Casbaneiro Banking Trojan Targets Financial Institutions in Argentina, Peru, Colombia and Mexico
Fortinet researchers identified a Casbaneiro campaign in August that specifically targets bank customers across four Latin American countries. The infection begins with a PDF attachment that displays the recipient's own email address to build credibility and creates urgency around an unpaid invoice or judicial notice. The PDF link performs IP-based geofencing, redirecting non-target visitors to Google or YouTube while delivering a Base64-encoded ZIP only to victims in the selected countries. Inside the archive, an HTA file downloads the legitimate AutoIt interpreter along with a compiled script and compressed payload, helping evade binary-focused defenses. The malware stays dormant until the victim visits a monitored banking website, at which point it activates its C2 channel, exfiltrates Outlook contact data, and can display bank-specific credential-harvesting overlays. Additional remote-access capabilities allow operators to control the keyboard, manipulate the clipboard, and execute arbitrary commands on the infected system.
Android Ransomware Mantax Otax Encrypts Files and Streams Victim Screen in Real Time
Researchers at Zimperium have identified a new Android ransomware strain called Mantax Otax that combines file encryption with live screen recording and surveillance capabilities. The malware is distributed through sideloaded APK files delivered via messaging apps and phishing messages rather than official app stores. Once installed, it uses AES encryption to lock files with the .enc extension and conducts ransom negotiations directly on the infected device through an on-screen chat interface. In addition to encryption, Mantax Otax continuously captures the screen, saves footage as MP4 video, and streams it to operators while also photographing the surroundings with the device camera. The malware abuses accessibility services to read on-screen content, intercepts SMS messages to steal two-factor authentication codes, and exfiltrates contacts, call logs, browser history, and credentials from WhatsApp and Telegram. It also displays a fake lock screen to capture the device PIN. Impact is reduced on Android 10 and later due to Scoped Storage restrictions, though surveillance functions remain active. Evidence points to a targeted campaign against users in Indonesia.
Russian State-Linked Group GTG-20006 Uses Anthropic AI Agents to Automate Malware Rebuilding
Anthropic has identified a Russian state-linked operation tracked as GTG-20006 that deployed autonomous AI agents to continuously rebuild its malware arsenal whenever detections occurred. The group, connected to Midnight Blizzard, APT29 and Cozy Bear, created a closed-loop automation system in which AI agents monitored tool performance against known defenses and triggered immediate code modifications to evade security products. Beyond malware, the agents handled domain registration, hosting infrastructure setup, phishing email delivery, command-and-control channel monitoring and implant persistence tracking across compromised environments. The campaign, active in July and August 2026 and overlapping with CaptiveCrunch, targeted more than twenty organizations including ministries, defense bodies, embassies and think tanks across Ukraine, Europe, the Middle East and Asia. In one incident the attackers exfiltrated over 300,000 national identity records and commercial registration data for more than 500,000 companies. Anthropic disrupted the activity and published a detailed report highlighting how the automation shifted the cost burden back onto defenders.
Check Point Patches Two Critical VPN Vulnerabilities Allowing Unauthenticated Remote Code Execution
Check Point has released fixes for two critical vulnerabilities affecting its Security Gateway, Security Management Server, and locally managed Spark Firewall products. Both flaws received a CVSS score of 9.8 and enable remote code execution without requiring authentication. CVE-2026-85102 resides in certificate trust validation during VPN negotiation, allowing an unauthenticated attacker to force the process and execute arbitrary code. CVE-2026-85103 is a heap buffer overflow triggered while parsing the ASN.1 structure of VPN certificates. Affected versions include R81.20, R82, and R82.10 before the listed hotfixes, as well as unsupported R80.40 and R81 releases. R82.20 remains unaffected. The issues were discovered internally by Check Point researchers, with no active exploitation or public proof-of-concept observed as of the September 10 disclosure.
Cybercriminals Distribute NJRAT, DCRAT and Chaos via Fake GTA 6 Downloads
Threat actors are leveraging anticipation around GTA 6 to spread multiple malware families through fake game downloads. Security researchers at Huntress identified campaigns that combine search-engine poisoning, gaming forums, torrent sites and social-media posts to deliver oversized fake ISO files exceeding 100 GB. Victims who execute the installer see Russian-language messages claiming an invalid crack or missing license, while remote-access tools and data stealers run silently in the background. The delivered payloads include NJRAT and DCRAT for keystroke logging, screen capture and webcam access, Mercurial Grabber for harvesting browser credentials and Discord tokens, and the Chaos wiper that encrypts small files and overwrites larger ones. The operation primarily targets Russian-speaking gamers, as indicated by the ransom note and error messages. Analysts recommend avoiding unofficial downloads and isolating any compromised systems immediately.
Microsoft Fixes Remote Code Execution Flaw in Windows Remote Desktop Client
Microsoft has addressed a remote code execution vulnerability in the Windows Remote Desktop Client tracked as CVE-2026-69485 with a CVSS score of 8.8. The flaw stems from the use of an uninitialized resource and can be triggered by a specially crafted network request from an authenticated attacker with low privileges. Exploitation requires no user interaction such as clicking links or opening files and can lead to information disclosure, file modification, or service disruption. Multiple versions of Windows 10, Windows 11, and Windows Server 2016 through 2025 are affected, including Server Core installations. The issue was not publicly disclosed prior to patching and no active exploitation has been observed. Fixes were released in the September security updates including KB5124008 and KB5122871.
US Accuses Chinese AI Companies of Industrial-Scale Model Distillation Targeting Claude, GPT, Gemini and Grok
US agencies have accused six Chinese artificial intelligence firms of conducting large-scale unauthorized distillation operations to replicate advanced capabilities from leading models including Claude, GPT, Gemini, and Grok. The activity is reported to have begun at least by late 2024 and involved DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun, and Z.AI. Billions of tokens were extracted through millions of automated API requests routed via cloud providers, aggregators, and proxies to conceal origins and evade detection. The targeted capabilities included chain-of-thought reasoning, programming, software engineering, autonomous agent functions, and multimodal processing. Shared premium accounts and bulk subscriptions were used to lower costs while automated route-switching systems helped maintain access after blocks. Authorities assess that the sophistication and volume indicate distillation has become a core development method for these companies and likely occurred with Chinese government awareness. China has rejected the claims, stating its AI progress stems from independent innovation and calling the allegations unfounded.
US Offers $10 Million Reward for Iranian IRGC Cyber Commander Amir Yaryab
The United States has announced a reward of up to $10 million through the Rewards for Justice program for information leading to the identification or location of Amir Yaryab, leader of the Cyber Operations Command within Iran's Islamic Revolutionary Guard Corps (IRGC). Yaryab oversees units responsible for cyber operations targeting critical infrastructure across the United States, Europe, and the Middle East. Groups under his direction, including Shahid Hemmat and Shahid Shushtari, have conducted campaigns against defense, energy, telecommunications, finance, transportation, hotels, and airlines sectors. He is also linked to structures associated with the CyberAv3ngers group, known for attacks on industrial control systems and operational technology equipment. Previous operations attributed to IRGC-linked actors compromised internet-exposed Unitronics programmable logic controllers, affecting at least 75 devices between November 2023 and January 2024, including 34 in the US water and wastewater sector. The reward specifically targets individuals acting under foreign government direction in malicious cyber activities against US critical infrastructure.
AI Detection Agents Shift Cybersecurity from Alerts to Autonomous Investigations
AI-powered detection agents are redefining how organizations identify and respond to cyber threats by moving beyond simple signature matching and alert generation. Instead of forwarding isolated events to human teams, these agents now perform initial investigations themselves, correlating telemetry from multiple sources and reconstructing attack chains before escalating any incident. The technology has progressed from theoretical discussions to active budget conversations, forcing security teams to evaluate how much triage work can be handled autonomously. Traditional monitoring systems deliver chronological events and leave analysts to connect the dots manually, while the new agents handle correlation, timeline verification, and scope assessment upfront. Attack automation has increased the urgency, as malicious actions can now produce effects faster than human analysis can complete. The agents collect artifacts, review asset history, check for repeated behavior, and determine the real impact, redistributing analyst workload without removing the need for human judgment on business impact and response decisions.
Critical ASUS Control Center Enterprise Flaw Allows Remote Root Access via CVE-2026-75754
A critical vulnerability identified as CVE-2026-75754 in ASUS Control Center Enterprise (ACC) carries a maximum CVSS score of 10.0 and enables unauthenticated remote attackers to gain full control of the management server and all connected devices. The flaw stems from a combination of missing authentication on a critical function, a server-side request forgery (SSRF) issue, and hardcoded credentials embedded directly in the software. Attackers can craft a malicious HTTP request to extract the system’s encryption key, activate an SSH service on TCP port 2222, and use fixed credentials to obtain a root shell without any user interaction. Once inside, the attacker can read, modify, or delete data stored in the ACC platform and propagate the compromise across managed servers, PCs, and workstations. All versions of ASUS Control Center Enterprise through 4.0.0.2 are affected. ASUS released a security advisory on September 4, 2026, urging immediate updates to mitigate the risk.
Critical Zero-Day 'StyleSmuggler' Vulnerability Exploited in Adobe Commerce and Magento
A zero-day vulnerability dubbed StyleSmuggler is being actively exploited in Magento Open Source and Adobe Commerce to achieve unauthenticated remote code execution and install backdoors on e-commerce servers. The flaw allows attackers to inject malicious PHP code into files generated by the platform and then force the template system to process it, with the attack chain triggered during the generation of default transaction failure emails. Exploitation has been observed since September 4 and works even if the email is not successfully sent. Researchers have reproduced the attack on clean installations of versions 2.4.7, 2.4.8, and 2.4.9, as well as on a fully patched Magento 2.4.6-p15 system. After compromise, a Rust-based implant is deployed outside the store directory, masquerading as the process [kworker/u:8:0] and maintained via a cron job that restarts it every five minutes. As of September 6, Adobe had not released a CVE, official patch, or specific workaround, though the next scheduled security update is set for September 8.
Pegasus Spyware Returns in Serbian Surveillance Campaign via Zero-Click iMessage Exploit
A Serbian student activist's iPhone was infected with the Pegasus spyware through a zero-click exploit in iMessage, allowing silent installation without any user interaction. The infection, confirmed by Citizen Lab in collaboration with the SHARE Foundation, showed indicators of compromise between December 2025 and January 2026. Apple later sent the target a notification warning of a mercenary spyware attack attempt. The exploit granted full access to photos, messages, files, and enabled covert microphone and camera activation. The vulnerability was addressed in the iOS 18.4.1 update released on April 16, 2025. The incident forms part of a wider surveillance wave in Serbia, with at least 14 individuals including students, activists, a parliament member, and a local political representative receiving similar Apple alerts. Additional targets were hit with Android spyware variants linked to NoviSpy.
Attackers Abuse Legitimate Node.js Runtime to Deploy Persistent Malware in Targeted Operations
Cybercriminals are increasingly misusing the legitimate Node.js runtime to execute malicious JavaScript scripts and maintain persistence on corporate networks. The technique has been observed since February 2026 against governments, technology firms, hotels, and financial organizations. Attackers favor the signed node.exe binary to evade signature-based detections after initial access via ClickFix social engineering. In one case against an Asian technology company, Node.js was deployed following blocked attempts to install AdaptixC2 and Cobalt Strike. The resulting implant establishes registry persistence and communicates with Ethereum blockchain gateways to retrieve commands hidden in smart contracts. The approach has been linked to multiple tools including ModeloRAT, Mistic, EtherRAT, and a variant of AsukaStealer, with some campaigns attributed to the initial-access broker Woodgnat, also known as KongTuke.
Dropbox Lenovo ID Integration Flaw Exposes Thousands of Accounts Without Passwords
A flaw in the federated authentication process between Dropbox and Lenovo ID allowed attackers to compromise approximately 5,000 accounts between August 4 and August 21, 2026. Attackers registered a Lenovo ID using the email address tied to an existing Dropbox account due to missing email verification checks. They then logged in via the Lenovo ID option, and Dropbox automatically associated the identity with the existing account without requiring the original Dropbox password. Only accounts linked to the Lenovo ID integration and lacking two-factor authentication were affected. In fewer than one-third of the compromised accounts, attackers viewed or downloaded stored files. Dropbox responded by terminating all Lenovo ID sessions, removing existing associations, and updating the login flow to require the Dropbox password. Lenovo described the issue as stemming from a legacy integration and stated that its own customers were not directly impacted.
Ransomware Operators Linked to The Gentlemen Deploy TukTuk C2 Framework for Espionage and Credential Theft
Operators associated with the ransomware group The Gentlemen have adopted a new command-and-control framework called TukTuk to steal credentials, monitor compromised systems, and prepare environments for ransomware deployment. The framework was discovered on a server that also hosted tools for disabling EDR solutions, research on vulnerable drivers, and data apparently stolen from two large organizations. TukTuk includes agents for both Windows and Linux, along with its own backend and management panel that allows remote command execution, file transfers, screen capture, and device tracking through a single interface. One notable feature displays a fake Windows Security window on the victim's machine to capture entered credentials and send them directly to the attackers' panel. Researchers identified a DLL sideloading technique that abuses the legitimate Greenshot.exe executable to load a malicious log4net.dll library and launch the TukTuk agent. The server also contained EDRKiller, WarsawKiller, and UnknownKiller tools, plus materials on BYOVD attacks that leverage vulnerable legitimate drivers to gain kernel access and interfere with security products.
Password Spraying Campaign Targets AWS Root Accounts in Over 150 Organizations
A password spraying campaign targeted AWS root accounts across more than 150 organizations between July 24 and August 23, 2026. Attackers performed repeated login attempts against identities holding maximum privileges in the cloud environment. The root account is created with every AWS account and grants full access to resources, configurations, billing, and sensitive administrative functions. Researchers observed a median of two attempts per organization, with some targets receiving up to eight attempts. No successful authentications linked to the campaign have been identified so far. The attacks leveraged distributed proxies across multiple countries and networks, including hosting infrastructure and residential proxies, while using user agents that mimicked older versions of Microsoft Edge and Firefox. Since June 2025, AWS has required MFA for root users, significantly raising the bar for account takeover even if a password is discovered.
Brazilian Malware BraZetsu Uses AI to Map and Sell Compromised Corporate Networks
Researchers have identified a new Windows malware called BraZetsu developed by the Brazilian actor Exilware. The tool maps compromised corporate networks, identifies high-value systems, and prepares access for sale to other criminals. The operation primarily targets Brazil and other countries in Latin America and the Iberian Peninsula. BraZetsu is written in Python with a modular architecture and maintains WebSocket communication with its command-and-control server. It searches for banking software, ERPs, e-commerce platforms, industrial environments, development tools, backup systems, and security solutions. The malware also looks for Brazilian CNAB financial files, digital certificates, and browser history. Evidence suggests heavy use of generative AI during development, with an AI component on the server that may classify stolen files and prioritize valuable machines.
Android 17 Adds Local Network Protection and Strengthens Wi-Fi Privacy Controls
Android 17 introduces Local Network Protection, requiring apps to obtain explicit permission before scanning or connecting to devices on the same Wi-Fi network. The feature limits unauthorized discovery of TVs, cameras, printers, consoles and other local equipment that could previously be used to build detailed user profiles. The update also enables Encrypted Client Hello by default to hide domain names during HTTPS handshakes from network observers. Certificate Transparency is now activated by default to detect fraudulent or mis-issued certificates that could enable interception attacks. Additional safeguards block forced downgrades to insecure 2G networks often exploited by fake base stations for SMS-based fraud. These changes collectively reduce passive tracking and man-in-the-middle risks without disrupting legitimate local network functions such as media casting.
Secure AI-Assisted Development: Five Critical Practices for Vibe Coding
AI has become the default path for quickly turning ideas into working applications, yet most security flaws in these projects stem not from model errors but from developer assumptions. The AI simply implements what is requested, and security requirements are rarely included in prompts. Five key areas account for the majority of issues: explicitly defining what an application must not do, distinguishing authentication from authorization, reviewing AI-suggested dependencies, removing secrets from commit history, and validating business logic rules that automated scans cannot detect. Applications built this way are already attracting cybercriminals because the resulting vulnerabilities can be identified and exploited at scale. The HackerSec Pentest Platform is positioned as an accessible option for vibe coders seeking quality security testing before production deployment.
Ransomware Group TITAN Deploys Local AI on AMD EPYC Servers to Accelerate Stolen Data Analysis
The TITAN ransomware group has announced the integration of an on-premises artificial intelligence platform designed to process up to 700 GB of exfiltrated data per hour. Operating as a ransomware-as-a-service model since May 2026, TITAN combines file encryption with data theft and has already published 24 victims across 10 countries. Manufacturing and professional services firms account for 29 percent of the targeted organizations. The AI system runs locally on AMD EPYC servers with GPU acceleration and automatically classifies financial documents, legal records, personal data, trade secrets, and intellectual property. It further identifies information with high reputational or regulatory impact, maps corporate and personal relationships, and estimates potential penalties under data-protection laws. The group also claims the platform can generate automated notifications to regulators and media outlets to intensify extortion pressure.
APT28 Expands Espionage with New HOOKEDGE Backdoor Targeting European Organizations
The Russian-linked APT28 group, also known as BlueDelta, has deployed a new lightweight backdoor called HOOKEDGE as part of a cyber-espionage campaign against strategic European entities. The attacks, assessed with moderate confidence, targeted government, diplomatic, and defense manufacturing organizations in Romania, Spain, and Turkey between September 2025 and April 2026. Infection begins with spear-phishing emails delivering Microsoft Word documents containing malicious macros that mimic official Spanish government materials. Upon execution, the macros drop files, establish persistence via scheduled tasks, and deploy the HOOKEDGE backdoor written in batch scripts. The malware uses hidden Microsoft Edge instances and the legitimate webhook.site service to blend command-and-control traffic with normal web activity. In high-value victims, operators installed a second HOOKEDGE instance with five-minute check-ins for faster control and data collection. The backdoor shows strong code similarities to the older HEADLACE implant previously attributed to the same group.
Iran-Linked Cyber Attack Leaves Small UK Power Plant Offline for Four Days
A cyber attack attributed to hackers with suspected ties to Iran took a small-scale UK power generation facility offline for approximately four days in July 2026. The incident affected a roughly 15 MW generator used to support peak demand periods, yet caused no customer outages or disruption to the national electricity grid. British authorities have not issued an official attribution, and investigators have not publicly identified the malware, vulnerability, or initial access vector used in the operation. Recovery required four days of extensive validation across controllers, configurations, security systems, and remote access points to ensure no residual risks remained. The case highlights the operational challenges of restoring industrial control environments after suspected nation-state activity. In response, UK authorities have strengthened guidance for the energy sector and are considering additional protective measures for critical infrastructure suppliers.
Critical Veeam ONE Flaw Allows Remote Attackers to Expose Service Account Credentials via Forced SMB Authentication
Veeam has patched a critical vulnerability in Veeam ONE that enables a remote unauthenticated attacker to force SMB authentication attempts using the platform's service account. Identified as CVE-2026-65641 with a CVSS score of 9.3, the flaw impacts version 13.1.0.7034 and all prior builds of the 13.x line, while 12.x installations remain unaffected. Exploitation allows an attacker to redirect SMB authentication to a controlled server, exposing Net-NTLM material that can be used for offline password cracking or NTLM relay attacks against other services. The severity depends on the privileges of the compromised account, with elevated risk in environments where Veeam ONE runs with high privileges or broad access to Active Directory, backup servers, and hypervisors. Patches are available in Veeam ONE 13.1 Patch 0 (build 13.1.0.7233) and Veeam ONE 13.0.2 Patch 1 (build 13.0.2.7159), and organizations are urged to apply updates immediately.
TeamViewer Patches High-Severity Flaws Allowing Remote Command Execution and Unauthorized File Writes
TeamViewer has addressed two high-severity vulnerabilities in its remote access clients that could enable attackers to execute commands and write files to unauthorized locations. The more critical issue, tracked as CVE-2026-19042 with a CVSS score of 8.8, affects Full Client and Host versions on Linux prior to 15.81.5 and permits command execution with the privileges of the logged-in user. Exploitation requires the attacker to be on the victim's contact list or for the victim to accept messages from external users. The second flaw, CVE-2026-16444 rated at CVSS 7.5, stems from improper file path validation and impacts Windows, macOS, and Linux clients before version 15.81.5, allowing path traversal sequences in transferred files or clipboard content during remote sessions. Legacy versions of TeamViewer 13, 14, and editions for Windows 7 and 8 also received targeted fixes. No active exploitation was observed at the time of disclosure, and the vendor urges immediate migration to the latest release.
Aurora Ransomware Affiliate Uses AI Assistant Cursor to Compromise Active Directory and VMware ESXi Servers
An affiliate of the Aurora ransomware group employed the AI-powered coding assistant Cursor to plan and execute targeted attacks against corporate environments, focusing on Active Directory and VMware ESXi servers. The campaign impacted more than 20 organizations between April and July 2026. The operator used Cursor to generate commands, refine exploitation techniques, and build attack sequences aimed at Windows infrastructure, particularly Active Directory Certificate Services (ADCS). Techniques included noPac exploitation combined with NTLM relay attacks leveraging PetitPotam, PrinterBug, and DFSCoerce to escalate privileges up to domain administrator level. After gaining access, the attackers collected and compressed large volumes of data for exfiltration before deploying the Aurora ransomware on Windows, Linux, and ESXi systems. The ESXi variant stops virtual machines prior to encryption to maximize impact on virtualized environments.
Next.js Patches Two Critical RCE Vulnerabilities in Versions 15.5.24 and 16.3.3
Next.js has released security updates to address two critical vulnerabilities that could allow unauthenticated remote code execution. The fixes are available in versions 15.5.24 and 16.3.3. The first issue, tracked as CVE-2026-75604 with a CVSS score of 9.0, affects applications hosted on Windows servers using Pages Router or App Router without Cache Components and stems from a path traversal flaw. The second vulnerability impacts the Image Optimization API when processing malicious AVIF files, enabling code execution through crafted image inputs. Both flaws affect a wide range of versions from 10.0.0 and 13.4 onward. Administrators are advised to update immediately, rebuild containers, and verify production environments run the patched releases, especially on Windows systems and those handling user-uploaded images.
Iran-Linked Tortoiseshell Group Deploys Malicious wtsapi32.dll Backdoor for Persistent Windows Access
Researchers have uncovered new tools deployed by the Tortoiseshell group, an Iranian-linked threat actor also tracked as Mirage Kitten, UNC1549, and Nimbus Manticore. The campaign features a Windows backdoor disguised as the legitimate wtsapi32.dll library that silently establishes reverse SSH tunnels over port 443 to maintain access to compromised networks. Active since at least 2018, the group has targeted defense, aerospace, technology, IT services, and military organizations primarily in the Middle East and the United States. The malware preserves expected Windows API functions while enabling command execution, file exfiltration, in-memory DLL loading, directory listing, and system reconnaissance. Associated infrastructure spans the United Arab Emirates, Saudi Arabia, the United Kingdom, Belgium, Canada, Australia, and Japan.
SonicWall Patches Critical Path Traversal and Update Flaws in NetExtender for Linux
SonicWall has released fixes for two high-severity vulnerabilities in its NetExtender client for Linux that could allow remote attackers to write arbitrary files with root privileges and manipulate the automatic update process. The flaws impact versions 10.3.5 and earlier, while the Windows version remains unaffected. CVE-2026-66152 carries a CVSS score of 8.8 and stems from improper handling of tar archives containing OPSWAT data, enabling path traversal that lets attackers escape the intended extraction directory. CVE-2026-66153 scores 7.0 and arises from inadequate symlink and temporary file handling in the NEService update mechanism. Both issues require user interaction to exploit, and no in-the-wild attacks have been observed so far. Administrators are urged to upgrade immediately to version 10.3.6 or later, as no workarounds exist.
Zscaler Fixes Multiple Critical Flaws in Client Connector Enabling RCE and Authentication Bypass
Zscaler has released patches for several vulnerabilities in its Client Connector agent that could lead to remote code execution, authentication bypass, local privilege escalation, and denial of service. The most severe issue, tracked as CVE-2026-59568 with a CVSS score of 9.1, allows unauthenticated remote attackers to execute arbitrary code within the context of the Zscaler Client Connector process. A second critical flaw, CVE-2026-59564 also rated CVSS 9.1, affects communication between the connector and its management portal, enabling attackers to circumvent authentication mechanisms. Additional vulnerabilities include a local buffer overflow on Android and ChromeOS tracked as CVE-2026-59566 with CVSS 8.4, as well as issues that could result in privilege escalation or service disruption. The flaws impact Client Connector versions across Windows, macOS, Linux, iOS, Android, and ChromeOS, with varying affected builds depending on the platform. Updated builds for lines 4.6 through 4.9 on Windows and equivalent fixes for other operating systems are now available.
WhatsApp Adds Support for Multiple Passkeys on Single Account
WhatsApp has expanded its account protection features by allowing users to register more than one passkey on the same profile. The update particularly benefits users who switch between Android and iOS devices while maintaining phishing-resistant login methods. Passkeys replace traditional codes and passwords with biometric authentication, facial recognition, or device lock mechanisms. Meta reports that more than one billion people already use the feature on WhatsApp, with Android support introduced in 2023 and iOS support added in 2024. The change reduces reliance on a single device during authentication by associating multiple keys with one account. Users can manage keys through Settings > Account > Passkeys. WhatsApp has also strengthened two-factor verification by allowing replacement of the six-digit PIN with longer passwords containing letters, numbers, and special characters.
AmnesiaStealer Malware for macOS Hijacks Authenticated Browser Sessions via ClickFix and Chromium
A new macOS malware strain named AmnesiaStealer has emerged that combines infostealing, persistence mechanisms, and the ability to silently take over already authenticated browser sessions. The infection chain begins with a fake GitHub page that tricks victims into copying a malicious command into Terminal using the ClickFix social engineering technique. Once executed, the Rust-based payload collects system data, displays a fake installer window to capture the user password, and attempts to unlock the Keychain for protected credentials. Targeted data includes browser cookies, history, extensions, documents, Apple Notes, Telegram sessions, and cryptocurrency wallet information. A secondary module copies the browser profile and launches a hidden Chromium instance controlled through the Chrome DevTools Protocol, allowing attackers to interact with active sessions that have already passed multi-factor authentication. Persistence is achieved via a disguised LaunchDaemon, and the malware attempts to remove installation artifacts afterward.
HackerSec's Yaga Pentest Agent Reaches 98.8% Effectiveness in White Box Testing
The offensive cybersecurity firm HackerSec announced that its Yaga pentest agent achieved a record 98.8% effectiveness in white box scenarios on the latest YagaBench evaluation. The agent also recorded 96.2% success in black box and 97% in gray box testing, marking the highest results since measurements began. These figures indicate that Yaga identified more than 98% of existing vulnerabilities across tested environments. The benchmark specifically highlights the performance gap between standalone AI models and the same models integrated into HackerSec's specialized pentest harness. Without the harness, models such as Opus 5 reached only 61% in white box testing, while GPT 5.6 SOL scored 60.9% in white box and 39.5% in black box. Yaga orchestrates four models during a single run, preserving context across phases and chaining findings to confirm exploitability while keeping false positives below 1%. CEO Andrew Martinez stated the company aims to reach 99% effectiveness across all pentest modalities by year end.
LockBit Claims Breach of U.S. Bank, Sets September 2026 Deadline for Ransom Payment
The ransomware group LockBit has added U.S. Bank to its leak site, claiming to have stolen data from the financial institution and threatening to publish it unless a ransom is paid by September 3, 2026. U.S. Bank stated it is actively investigating the claims but has so far found no evidence of unauthorized network access or impact on internal systems. Lee Henderson, the bank's vice president of public affairs, confirmed that the institution is monitoring the situation and treating the allegations seriously. LockBit placed the bank's domain on its extortion portal on August 19 and has a history of resuming operations after a major international law enforcement disruption in 2024. The investigation remains ongoing, with the final scope depending on whether unauthorized access and data theft can be confirmed.
Chinese Cybercriminals Deploy AI Agents to Automate Exploitation of Exposed Windows and Linux Servers
A Chinese cybercriminal group tracked as UAT-10147 is actively using AI agents to automate reconnaissance, exploitation, and post-exploitation against internet-facing Windows and Linux servers. Cisco Talos identified the campaign after discovering approximately 170,000 URLs in the attackers' infrastructure, with targets spanning government, education, media, technology, and gaming organizations across multiple countries including Brazil. The group divides targets into smaller batches to accelerate scanning and exploitation of known vulnerabilities in applications such as Zimbra, AjaxPro, Nacos, and Telerik. AI agents go beyond simple code generation by producing exploitation guides, refining failed attacks, verifying payload success, and automating persistence and post-exploitation steps. In one observed case, the agents created scripts capable of validating permissions, exploiting ViewState deserialization, installing implants, and deploying web shells on compromised IIS servers. Cisco Talos assesses the operation as a shift toward semi-autonomous offensive orchestration rather than basic AI-assisted scripting.
AWS Details Architecture to Reduce Prompt Injection Risks in AI Agents
AWS has introduced a new architecture designed to prevent compromised or manipulated AI agents from accessing data beyond user permissions. The approach relies on Amazon Bedrock AgentCore to shift authorization decisions from the agent itself to the underlying infrastructure and connected services. The core risk arises when agents receive broad credentials to query databases, repositories, and SaaS platforms, allowing potential prompt injection attacks to retrieve unauthorized information. In the proposed design, users authenticate via Amazon Cognito and receive JWT tokens containing attributes such as department or role. The AgentCore Runtime validates these tokens before executing any agent actions, rejecting requests that violate configured rules. For DynamoDB queries, temporary credentials are issued through AssumeRoleWithWebIdentity, with IAM policies enforcing strict access to authorized data partitions only.
Manic Android Malware Steals PINs via Transparent Overlay and Relays Data Through Nearby Infected Devices
A newly identified Android malware strain named Manic merges banking trojan, spyware, and remote access capabilities. The threat has been active since at least February 2026 and continues to receive updates that add anti-analysis features, in-memory code loading, and lock-screen credential theft. Manic monitors 169 financial, messaging, and government applications while using a transparent overlay on legitimate numeric keyboards to capture PINs without displaying a full fake banking screen. Stolen data can be forwarded through other compromised nearby devices even when the original phone lacks internet connectivity. Operators also leverage WebRTC sessions for live screen viewing and remote interaction. The malware additionally functions as a keylogger, intercepts SMS and notifications, and collects passwords, one-time codes, and recovery phrases. Security researchers recommend avoiding unknown APKs and scrutinizing requests for Accessibility Services or broad device control permissions.
Oracle Issues Emergency Patches for Critical Remotely Exploitable WebLogic Server Vulnerabilities
Oracle released an emergency security update on August 18 containing 943 new patches across dozens of enterprise product families. The update addresses multiple high-severity flaws in Oracle WebLogic Server that can be exploited remotely without authentication over IIOP, T3, and RMI protocols. Four vulnerabilities—CVE-2026-60698, CVE-2026-60672, CVE-2026-60696, and CVE-2026-60977—received CVSS scores of 9.8 and can impact confidentiality, integrity, and availability of affected servers. A separate critical issue, CVE-2026-61241, was disclosed in the LDAP server component of Oracle Internet Directory with a maximum CVSS score of 10.0 and affects versions 12.2.1.4.0 and 14.1.2.1.0. Oracle urges customers to maintain supported releases and apply patches promptly, noting prior incidents where unpatched systems were targeted after fixes became available. Organizations are advised to validate patches in test environments before production deployment to minimize operational risk.
Google Patches Two Critical Memory Corruption Flaws in Chrome WebGL and Dawn Components
Google has issued a security update for Chrome that addresses 15 vulnerabilities, two of which are rated critical. The flaws, tracked as CVE-2026-76034 and CVE-2026-76036, involve buffer overflow conditions that can lead to out-of-bounds memory writes. CVE-2026-76034 affects the WebGL component used for 2D and 3D graphics rendering on web pages, while CVE-2026-76036 impacts Dawn, the Chromium implementation of WebGPU. Both issues were discovered internally and could result in crashes or remote code execution in certain scenarios. Updated versions are now available for Windows, macOS, Linux, and Android, and users are strongly advised to apply the patches immediately.
Critical Vulnerability in Forminator Forms WordPress Plugin Enables Unauthenticated Remote Code Execution
A critical vulnerability tracked as CVE-2026-15748 with a CVSS score of 9.8 has been identified in the Forminator Forms plugin for WordPress. The flaw allows unauthenticated attackers to upload malicious PHP files by manipulating the Select field alongside a File Upload field on published forms. This bypasses extension blocking and file type validation, potentially leading to full server compromise including web shell installation and database access. The plugin, which has more than 600,000 active installations, is affected in all versions up to and including 1.56.1. The issue was resolved in version 1.56.2 released at the end of July, with later updates including 1.57.0 now available.
Snowflake GitHub Actions Workflow Flaw Enabled Command Injection and Jira Token Theft
A vulnerability in Snowflake's GitHub Actions workflow in the snowflakedb/snowflake-connector-net repository allowed any user to execute arbitrary commands on a runner by creating a specially crafted issue in a public repository. The flaw stemmed from unsanitized insertion of issue titles and content directly into shell commands, combined with a failed access control check that relied on a non-existent pull request property during issue events. During authorized testing, an autonomous security agent exploited the issue to extract a Jira API token from the pipeline environment variables. This credential granted read access to internal engineering, security compliance, and bug bounty projects. The vulnerability was active from June 18 to June 23, 2026. Snowflake patched the workflow on the day the report was received and rotated the compromised token the following day. The vulnerable code appeared in a pull request that involved GitHub Copilot, though available history does not confirm the AI generated the insecure lines.
Apple Patches CVE-2026-43760 Screen Sharing Flaw Granting Root Access on macOS
A vulnerability in the macOS Screen Sharing feature allows remote attackers to execute commands with root privileges under specific configurations. Identified as CVE-2026-43760 and tied to the screensharingd service, the flaw affects systems with Screen Sharing or Remote Management enabled alongside the legacy VNC password option. Attackers who know the VNC password can exploit unauthenticated account binding to retrieve protected files such as /etc/sudoers or write policies into /private/etc/sudoers.d. This grants non-privileged accounts passwordless sudo rights. The issue stems from VNC-authenticated connections not being mapped to specific macOS accounts while file-transfer components retain root privileges. Apple addressed the vulnerability in macOS Tahoe 26.6 and macOS Sonoma 14.8.8, both released on July 27, 2026.
Cybercriminals Deploy Advanced AI for Continuous Automated Reconnaissance and Exploitation at Scale
Advanced AI models now enable cybercriminal groups to maintain uninterrupted reconnaissance across enterprises in every sector, mapping domains, exposed services, and infrastructure changes at a pace no human team could sustain for weeks. The automated process targets two opposite profiles of vulnerable systems: legacy environments left unpatched for years with outdated versions and forgotten permissions, and rapidly deployed applications built through Vibe Coding that reach production without security review. The entire attack chain—reconnaissance, vulnerability identification, validation, and exploitation—is now executed by AI agents operating with minimal human oversight. What previously required a dedicated specialized team focused on one target at a time now runs in parallel against thousands of targets simultaneously, with marginal cost approaching zero for each additional attack. This collapse in operational costs and rise in success rates has restructured the cybercrime economy, funding increasingly sophisticated tools and lowering the barrier for new operators who need only platform access rather than deep technical expertise. The time window between an application reaching production and discovery by attackers has shrunk to hours, while most organizations still treat security as a finite project rather than an ongoing process.
Microsoft to Make Passkeys Default Authentication in Entra ID Starting September 2026
Microsoft will transition Entra ID users to passkeys as the standard authentication method beginning September 1, 2026. The change aims to eliminate reliance on phishing-prone SMS and voice call codes. Existing SMS and voice users will be prompted to register passkeys during their next multifactor authentication. Passkeys rely on public-key cryptography and avoid shared secrets, blocking phishing, interception, credential reuse, and SIM swapping attacks. Support includes synced passkeys via iCloud Keychain and Google Password Manager, plus device-bound options like Microsoft Authenticator, Windows, and FIDO2 hardware keys. Native SMS and voice services will be fully retired on February 1, 2027, forcing affected tenants to adopt passkeys with no opt-out option.
Dysphoria Botnet Compromises Nearly 300,000 Devices for DDoS Attacks and Residential Proxy Services
The Dysphoria botnet has infected approximately 296,000 devices, including routers, IP cameras, gateways, and embedded Linux systems. Researchers first observed the threat in the first quarter of 2026, noting rapid evolution from the jackskid and fbot families. The infrastructure is primarily used for DDoS attacks but has expanded to offer residential proxy capabilities. Infection occurs through brute-force attacks on Telnet and SSH services with weak credentials, as well as known remote code execution vulnerabilities in IoT equipment. A key technical advancement involves the use of Ethereum and Solana blockchain domains for command-and-control infrastructure, making takedowns significantly harder. Compromised devices can also function as relays by leveraging UPnP to expose ports and hide criminal traffic origins. Operators advertise attack capacity of up to 4 Tbps and sell DDoS services in structured commercial packages targeting internet services and gaming platforms worldwide.
Fortinet Patches Critical Authentication Bypass in FortiWeb Allowing Login with Random Credentials
Fortinet has released security updates addressing multiple vulnerabilities across FortiWeb, FortiManager, and FortiClient products. The most severe issue, tracked as CVE-2026-26035, affects FortiWeb and carries a CVSS score of 8.8. The flaw occurs when administrative accounts configured with remote RADIUS authentication use the wildcard option, which is disabled by default. Under these conditions, an unauthenticated remote attacker can gain administrative access to the web application firewall by supplying arbitrary usernames and passwords. Successful exploitation grants full control over the FortiWeb instance, potentially compromising web application security. The vulnerability impacts FortiWeb versions 8.0, 7.6, 7.4, 7.2, and 7.0, with fixes available in versions 8.0.3, 7.6.7, 7.4.12, and 7.2.13. No active exploitation of the authentication bypass flaws has been observed so far, but the risk of remote administrative access makes immediate patching essential for exposed appliances.
Microsoft Patches Seven Exchange Server Vulnerabilities Including Critical Remote Code Execution Flaw
Microsoft has addressed seven vulnerabilities in Exchange Server that enable remote code execution, privilege escalation, denial of service, content spoofing, and security feature bypass. The most severe issue, CVE-2026-62913, carries a CVSS score of 8.8 and involves a heap buffer overflow that can be exploited remotely by a low-privileged attacker without any user interaction. Successful exploitation grants code execution on the server, facilitating email theft, persistence mechanisms, lateral movement, and ransomware deployment. CVE-2026-62911, demonstrated at Pwn2Own Berlin with a CVSS score of 8.0, allows authentication bypass through capture and replay of credentials. Additional fixes cover SSRF-based privilege escalation, remote deserialization crashes, content spoofing, and authorization bypasses. Patches are available for Exchange Server Subscription Edition, while 2016 and 2019 versions receive updates only through the Extended Security Update program.
US Presidential Memo Authorizes Selected Private Companies to Join Federal Cyber Operations Against Foreign Criminal Groups
The United States government has established a formal program allowing vetted private-sector companies to participate in offensive cyber operations targeting foreign criminal organizations. Signed by President Donald Trump on August 12, 2026, the presidential memorandum places the initiative under the National Coordination Center with joint oversight from the Department of Justice and the Department of Homeland Security. Participating firms will operate exclusively under government contracts, direction, and supervision, with strict requirements including technical evaluations, financial guarantees of at least one million dollars, and pre-approval for every operation. The program focuses on disrupting ransomware, phishing, financial fraud, and other schemes affecting American citizens while imposing clear limits to prevent unintended harm to US persons or escalation to prohibited levels of force. In contrast to Brazil’s ongoing policy discussions, the US move formally recognizes that advanced offensive capabilities now reside primarily in the private sector and creates a regulated mechanism to access them. Operational rules must be published within 60 days, marking a significant shift in how governments integrate private expertise into state-directed cyber actions.
Microsoft Patches High-Severity Outlook Flaw Allowing Remote Code Execution
Microsoft has addressed a high-severity vulnerability in Outlook tracked as CVE-2026-70329 with a CVSS score of 8.8. The flaw stems from an integer overflow during data processing that can trigger memory corruption and alter the application's execution flow. Exploitation occurs over the network without requiring authentication or privileges, though some user interaction is necessary to process the malicious content. Successful attacks grant code execution under the privileges of the logged-in user, potentially enabling malware installation, file access, data theft, or lateral movement. The issue impacts Microsoft 365 Apps for Enterprise, Office 2019, Office LTSC 2021, Office LTSC 2024, and Outlook 2016 on both 32-bit and 64-bit systems. A dedicated patch KB5002755 updates Outlook 2016 to build 16.0.5565.1000, while Click-to-Run versions receive fixes through standard Microsoft update channels.
Lazarus Group Exploits Windows Kernel Zero-Day CVE-2026-68820 to Deploy FudModule Rootkit
The North Korean Lazarus APT group has been actively exploiting a zero-day vulnerability in the Windows kernel to escalate privileges to SYSTEM level and install the FudModule rootkit. The flaw, tracked as CVE-2026-68820, resides in the afd.sys driver responsible for network functions and socket management. Microsoft released a patch for the issue on August 11. The attacks form part of the ongoing Operation Dream Job campaign, which uses fake job offers to target professionals in defense, aerospace, and aviation sectors. Victims in Brazil, Europe, and India are tricked into opening malicious PDF viewers or prepared files that deliver the MISTPEN downloader. Once initial access is obtained, the zero-day exploit elevates privileges, allowing FudModule to tamper with Windows telemetry and weaken EDR solutions as well as Smart App Control.
SAP Releases August Security Update Patching 28 Vulnerabilities Including Critical CVSS 10 Flaw
SAP has issued a broad security update to address multiple critical vulnerabilities that could enable code injection, memory corruption, and privilege escalation across enterprise systems. The August package includes 28 new security notes along with a GitHub advisory and two prior fix updates. The highest-severity issue, CVE-2026-58231, carries a maximum CVSS score of 10 and affects the Data Hub Adapter in SAP Commerce Cloud versions 2211 and 2211-JDK21, allowing remote exploitation without user interaction due to improper authorization. Another critical flaw, CVE-2026-44772 rated 9.9, impacts SAP Manufacturing Integration and Intelligence 15.4 and 15.5, permitting malicious code injection into industrial process monitoring systems. Additional vulnerabilities rated 9.1 and lower cover directory traversal, SQL injection, XSS, XXE, hardcoded credentials, and OS command injection across various corporate components. Administrators are advised to identify affected systems and prioritize installation of the critical patches first.
Zoom Patches Zoomsday Vulnerability Enabling Remote Code Execution in Meetings
Zoom has addressed four vulnerabilities that could allow attackers to compromise meeting participants, including flaws leading to remote code execution without any victim interaction. The most severe issue, CVE-2026-53413, rated 8.3 and nicknamed Zoomsday, resides in the annotation feature used for drawing, highlighting, or adding text during screen sharing. This component processes network data using fixed 128-byte buffers without proper size validation, enabling memory corruption that alters program execution flow. Researchers demonstrated the attack on macOS by silently launching Safari on the victim's machine. The flaws affect Zoom Workplace, Zoom Rooms, Meeting SDK, and VDI clients. Users must update to patched versions such as Workplace 7.1.5 or 7.0.6, and Zoom Rooms or Meeting SDK 7.1.5 to mitigate annotation-related risks. No active exploitation has been observed publicly, yet centralized enterprise deployments require priority updates since attacks can occur during live meetings.
Ransomware Groups Disable EDR, Backups and Windows Telemetry Before Encryption
Ransomware operators are increasingly focusing on disabling endpoint detection and response tools, backup systems, and Windows telemetry mechanisms prior to launching encryption. An analysis of the ten ransomware families with the lowest prevention rates in 2026 found that Play achieved only 13 percent of attacks blocked. BlackByte followed with 25 percent blocked and LockBit with 30 percent blocked. BabLock leverages a legitimate uninstaller to remove endpoint protection and terminates processes belonging to antivirus, EDR, backup, and database applications. It then clears the Security and System event logs to hinder incident response. LockBit 5.0 instead interferes with Event Tracing for Windows to reduce visibility for monitoring solutions. Additional families employ process injection, in-memory execution, registry modifications, file masquerading, and living-off-the-land binaries to evade detection.
Google Play Apps Hide Anatsa Banking Trojan Using Stealthy Loaders
Malicious applications distributed through Google Play are employing stealthy loaders to install the Anatsa banking trojan for Android, putting victims' financial accounts at risk. The campaign relies on seemingly legitimate programs to conceal the initial infection stage, with one identified case involving a modified PDF reader that displays a fake update prompt to trick users into installing an additional component. Another loader discovered in the Cleanova application collects data from analytics SDKs, including installation source details, and forwards this information to a command-and-control server. The server then decides whether to deliver the malicious payload based on predefined criteria, leaving the code inactive for non-targeted victims to evade automated analysis. Once Anatsa is installed, the threat can facilitate theft of credentials used for financial services and enable fraudulent transactions. Banking trojans represented 30.77 percent of detected malicious mobile applications in the second quarter of 2026, during which more than 1.99 million attacks involving malware, adware, or unwanted programs were blocked and 93,574 banking trojan installation packages were identified.
WordPress Patches High-Severity XSS2Shell Flaw Enabling Remote Code Execution (CVE-2026-64638)
WordPress has released a fix for a high-severity vulnerability in its core that can turn a malicious login attempt into remote PHP code execution on the server. The flaw, tracked as CVE-2026-64638 and rated 8.9 on CVSS, is named XSS2Shell and begins as a reflected cross-site scripting issue on the authentication screen. An unauthenticated attacker can submit a specially crafted username that bypasses sanitization mechanisms due to differences in how PHP and WordPress interpret certain characters, allowing controlled HTML elements to be injected into the login page. This leads to JavaScript execution in the site domain, and when an authenticated administrator visits a prepared page, the attacker can abuse the existing session to authorize application credentials and upload arbitrary PHP code. The patch is included in WordPress 7.0.3 released on August 6, with backports available for branches as old as 4.7. No public exploitation has been confirmed yet, but the technical details make immediate updating a priority for site administrators.
Google Releases Chrome 151 Fixing 41 Vulnerabilities Including Six Critical Flaws
Google has issued Chrome 151 to address 41 security vulnerabilities, six of which are rated critical and could lead to memory corruption, browser crashes, and remote code execution. The update covers Windows, macOS, and Linux platforms with versions 151.0.7922.108 and 151.0.7922.109. Two use-after-free flaws in the WebGL component, tracked as CVE-2026-19137 and CVE-2026-19170, allow continued memory access after deallocation. Additional critical issues affect the Aura, Skia, Views, and ANGLE components under CVE-2026-19149, CVE-2026-19154, CVE-2026-19172, and CVE-2026-19157. Google has withheld technical details and proof-of-concept code for several flaws to limit exploitation while the rollout completes. Users are advised to check for updates immediately through the browser's About section.
Cisco Patches Seven Critical IOS XE Vulnerabilities Including CVSS 9.8 Command Injection Flaw
Cisco has released security updates to address seven vulnerabilities in IOS XE, the operating system powering routers, switches, wireless controllers, and other enterprise network devices. The highest-severity issue, tracked as CVE-2026-20272 with a CVSS score of 9.8, stems from improper handling of special elements in commands and could enable remote command injection, unauthorized operations, and full device compromise over the network. Additional flaws impact memory management, resource lifecycle, numeric calculations, execution flow, and input validation, potentially leading to buffer overflows, out-of-bounds writes, crashes, infinite loops, and directory traversal. The vulnerabilities affect IOS XE versions 17.9, 17.12, 17.15, 17.18, and 26.1 in both standalone and controller modes regardless of configuration. Fixed releases include 17.9.10, 17.12.8, 17.15.6, 17.18.4, 17.18.4a, and 26.1.2, with no available workarounds. The issues were discovered during internal testing that incorporated advanced AI models, and no evidence of public exploitation has been observed.
AWS and Vercel Patch Critical Flaws in AI Agent Platforms Allowing Unauthorized Tool Execution
AWS and Vercel have addressed multiple critical vulnerabilities in their AI agent platforms that enabled unauthorized execution of tools without legitimate model approval. The issues, grouped under the CoreBreak pattern, allowed attackers to bypass AI authorization checks by injecting crafted tool calls that the infrastructure misinterpreted as model-approved actions. In AWS, CVE-2026-18830 affected the InvokeHarness API in Amazon Bedrock AgentCore, permitting authenticated users to trigger sensitive tools directly. Vercel faced two separate flaws tracked as CVE-2026-64650 and CVE-2026-64651 that let sandboxed code reach host system tools, potentially exposing secrets or cloud APIs. No public evidence of active exploitation has been confirmed yet. Organizations are advised to apply updates immediately, restrict available tools for agents, and treat all external inputs as potentially malicious.
Free Online Panel Examines Rising Omnichannel Scams and Multichannel Fraud Tactics
The Brazilian human risk management firm Eskive is hosting its third free online panel on August 18 at 11 a.m. to address the growing threat of omnichannel cyber fraud. Experts will discuss how attackers combine multiple channels such as email, SMS, and other vectors to create more convincing social-engineering narratives that bypass traditional single-channel defenses. The event will feature CEO Priscila Meyer as moderator along with cyber threat intelligence specialist Thiago Bordini and Santa Catarina Civil Police investigator Elias Edenis. Participants will gain practical insights from real client simulations, live Q&A sessions, and interactive quizzes designed to improve organizational preparedness. The panel aims to highlight why users accustomed to recognizing basic phishing or smishing attempts remain vulnerable when fraudsters deploy coordinated, multi-channel campaigns.
OpenAI Disables Coordinated ChatGPT Network Used for Financial Scams and Identity Forgery
OpenAI has deactivated a coordinated network of ChatGPT accounts that supported financial fraud, romance scams, and identity forgery operations. Criminals leveraged the AI to generate fake personas, translate conversations, and craft targeted messages aimed at victims across multiple schemes. The investigation originated from reports of suspicious activity observed on WhatsApp. Scammers used the tool to produce forged documents including stock confirmations, legal notices, passports, and fake financial interfaces to increase credibility. Operations typically began on social media or messaging apps, building emotional trust or urgency before requesting deposits, activation fees, or nonexistent fines. Indicators of possible human trafficking and forced labor were also uncovered through job advertisements and internal discussions about worker control in Poipet. OpenAI has blocked the accounts and shared operational indicators with law enforcement and technology companies.
Critical Django Vulnerability CVE-2026-15307 Enables Remote Code Execution via GDALRaster Spatial Queries
The Django project has released versions 6.0.8 and 5.2.17 to address four vulnerabilities that could result in arbitrary file writes, unauthorized requests, denial of service, and cross-site scripting attacks. The most severe issue, tracked as CVE-2026-15307, impacts spatial queries that process raster data through the GDALRaster interface. Attackers can supply specially crafted values that cause the Django process to write files or initiate external connections under its own permissions, potentially leading to remote code execution in certain configurations. Exploitation is also possible through the Django admin panel when a staff user has view permissions on models containing geographic fields. The patches have been backported to the 5.2 and 6.0 branches as well as the upcoming 6.1 release candidate. In addition to fixing the flaw, the updates alter query behavior to reject dictionaries and strings that do not represent valid geometries.
Microsoft Copilot Abused to Accelerate Corporate Account Takeovers and Bank Transfer Frauds
Cybercriminals can leverage Microsoft Copilot to speed up the compromise of corporate accounts and execute financial frauds. In a controlled simulation, researchers demonstrated how the AI assistant helped redirect a $247,500 bank transfer after gaining access to an employee's email. The attack begins with the creation of inbox rules that automatically delete suspicious login alerts, followed by rapid analysis of internal communications to identify high-value targets. Copilot was then used to replicate the victim's writing style and craft a convincing phishing email capable of stealing session tokens and bypassing multi-factor authentication. Once the executive account was taken over, the tool quickly located pending payments and generated a fraudulent request to alter the recipient bank details. The scenario highlights risks from legitimate AI features rather than any specific vulnerability in Copilot itself.
DarkSword Exploit Kit Expands to 180 Sites Targeting iPhone Users with Zero-Click Chain
The DarkSword exploit kit has scaled its infrastructure to 180 web properties hosted on 27 servers, enabling zero-click compromise of iPhones running vulnerable iOS versions. Attackers deploy fake pages impersonating Apple, AWS, and other popular services to trigger a six-vulnerability exploit chain that achieves remote code execution, sandbox escape, and privilege escalation on iOS 18.4 through 18.7. Once the device is compromised, GHOSTBLADE components extract data from the Keychain, iCloud files, Wi-Fi passwords, and other stored content before exfiltrating it to attacker-controlled panels. The malware deletes crash reports and forensic logs to hinder analysis and does not establish traditional persistence, limiting the window of data theft to a single visit. Infrastructure rotates rapidly, with five of seven administrative panels observed on July 30, 2026, having appeared only a week earlier, indicating frequent domain and host replacement to evade blocking.
Web Application Vulnerabilities Surge as Rapid Development Outpaces Security Reviews
The number of vulnerabilities in web applications continues to rise each quarter with no signs of slowing. Frequent releases, lean teams, and pressure for speed are pushing systems into production with flaws that should have been caught earlier. Vibe Coding is accelerating the problem by enabling anyone to launch applications in days and push dozens of updates daily without integrating cybersecurity into the process. Most new software reaches production without any security review, and attackers are already exploiting the common weaknesses these applications share. Companies face data leaks, service disruptions, and customer exposure from issues that a single test could have identified. The recommended response is AI-driven pentesting that simulates real attacker behavior instead of relying on known patterns. Launching web applications without security testing is becoming an increasingly expensive decision as incident costs far exceed preventive measures.
Critical Stack Buffer Overflow in TP-Link TL-WR940N Enables Remote Code Execution
A high-severity vulnerability tracked as CVE-2026-12935 with a CVSS score of 8.7 affects the TP-Link TL-WR940N router on hardware version V6. The flaw resides in the RTSP connection tracking module responsible for managing audio and video streaming sessions over the network. It is caused by a stack-based buffer overflow that allows oversized data to corrupt kernel memory, potentially leading to device crashes or full remote code execution. No administrative credentials are required for exploitation, though the attack depends on an RTSP connection initiated by a device already present on the local network. Successful compromise grants attackers the ability to alter router settings, modify DNS servers, intercept traffic, redirect users to malicious sites, and pivot to other connected devices. Users are advised to verify the hardware revision on the device label and apply the region-specific firmware update released by TP-Link.