Claude Opus 5, the latest flagship AI model from Anthropic, represents a significant shift in how advanced systems can be safely utilized in cybersecurity. This model can proactively identify software vulnerabilities while specifically preventing the generation of exploits and offensive…
Category: GBHackers Security | #1 Globally Trusted Cyber Security News Platform
Europol Launches Project COMPASS to Disrupt ‘The Com’ Cybercrime Network Targeting Minors
Europol has launched Project COMPASS, a coordinated transnational initiative aimed at disrupting “The Com,” a highly dangerous cybercrime and nihilistic extremist network that systematically targets minors and vulnerable young people across digital platforms. The Com operates as a sprawling transnational…
vBulletin Pre-Auth RCE Flaw Allows Remote PHP Code Execution
A critical pre-authentication remote code execution vulnerability in vBulletin, tracked as CVE-2026-61511, could allow unauthenticated attackers to execute arbitrary PHP code on vulnerable forum servers. This issue affects vBulletin versions 6.2.1 and earlier, as well as 6.1.6 and earlier, according…
Over 70 Fake Windows App Sites Could Turn Trusted Downloads Into Malware
A newly uncovered cluster of more than 70 impersonation domains targeting popular Windows applications is raising fresh concerns about a scalable malware distribution campaign that leverages trust in legitimate software ecosystems. The discovery, triggered by a developer investigating unusual search…
GitHub Adds Dependabot Cooldown to Stop Poisoned Dependencies
GitHub has introduced a default cooldown period for Dependabot version updates to decrease the risk of organizations automatically adopting malicious or compromised open-source dependencies as soon as they are released. This change comes in response to a rise in supply…
Crypto Criminals Use Social Media Profiling to Select Victims for Violent Wrench Attacks
Crypto criminals are increasingly weaponizing social media intelligence to identify and target high-value individuals in a surge of violent “wrench attacks,” marking a shift from purely digital exploitation to coordinated physical coercion campaigns. Recent threat intelligence indicates that attackers are…
Iranian Hackers Exploit Rockwell, Schneider and Siemens PLCs Across U.S. Critical Infrastructure
Iranian-affiliated advanced persistent threat (APT) actors are actively exploiting internet-connected programmable logic controllers (PLCs) from major industrial vendors, including Rockwell Automation, Schneider Electric, and Siemens, targeting U.S. critical infrastructure sectors. A joint cybersecurity advisory (AA26-097A) released by the FBI, CISA,…
Windows WalletService Flaw Lets Standard Users Gain SYSTEM Privileges
Microsoft Windows WalletService is affected by a local privilege escalation vulnerability tracked as CVE-2026-49176. This flaw could allow a standard authenticated user to obtain SYSTEM-level privileges. The vulnerability arises from WalletService’s handling of user-controlled file paths during initialization. An attacker…
New usbliter8 SecureROM Exploit Enables iOS 27 Jailbreak on iPhone 11 Pro
A new developer-focused project, usbliter8-fun, showcases an iOS 27 jailbreak workflow specifically for the iPhone 11 Pro. It combines the usbliter8 SecureROM exploit with a custom firmware restoration process. This proof of concept targets Apple’s A13-based iPhone 11 Pro. It…
Microsoft Introduces KMS Hardware-Secured for Windows Server Activation
Microsoft has introduced KMS Hardware-Secured, an upcoming enhancement to Windows Server activation that utilizes Trusted Platform Module (TPM)-based attestation to validate Key Management Service (KMS) hosts before they can activate Windows devices. Announced in a July 2022 Windows IT Pro…
Apple Biome Data Reveals Safari Activity, Wallet Transactions and Location Visits
Apple’s internal Biome framework is rapidly emerging as one of the most valuable sources of forensic intelligence on iOS, with newly analyzed data revealing detailed records of Safari browsing activity, Wallet transactions. Originally misunderstood due to its name, Biome is…
AWS Launches Claude Opus 5 With Advanced Agentic Coding and Cybersecurity Capabilities
AWS has launched Anthropic’s Claude Opus 5 on Amazon Bedrock and the Claude Platform, introducing a groundbreaking model designed for agentic coding, complex enterprise workflows, visual understanding, and long-running autonomous tasks. Released on July 24, 2026, Claude Opus 5 is…
Claude Code Symlink Flaw Exfiltrates Sensitive Files Without User Approval
Claude Code has a flaw in its startup memory loader related to handling symbolic links (symlinks) that can unintentionally expose readable files from outside a cloned repository, without requiring explicit user approval. The issue arises not from the tool following…
SparkKitty Monitors Mobile Photo Galleries and Exfiltrates Sensitive Images to C2 Servers
SparkKitty is a cross‑platform mobile stealer that weaponizes users’ photo galleries, using OCR to extract sensitive text from images and silently exfiltrating it to attacker‑controlled C2 servers on both Android and iOS. Built as an apparent successor to SparkCat, the…
PyPI Blocks New File Uploads to Old Releases to Prevent Package Poisoning Attacks
PyPI has introduced a new supply-chain security control that prevents publishers from uploading additional files to package releases older than 14 days, reducing the risk of attackers poisoning previously trusted versions after compromising project credentials, automation workflows, or publishing tokens.…
BlueNoroff Fake Meeting Kit Captures Webcams, Disables Defender and Steals Cryptocurrency Credentials
BlueNoroff, a financially motivated threat cluster linked to the Lazarus Group, has been observed deploying a highly sophisticated “fake meeting” phishing kit. That goes far beyond traditional lures, enabling webcam capture, Microsoft Defender evasion, and targeted cryptocurrency credential theft. New…
Google Indexed Claude AI Shared Chats Exposing Sensitive User Conversations
Anthropic’s Claude includes a share feature that creates a publicly accessible URL for conversations, allowing users to share AI chats with colleagues, clients, or friends. However, this convenience also brings exposure risks when shared URLs are posted in public forums,…
GitLab RCE Flaws Allow Attackers to Execute Commands via Malicious Jupyter Notebooks
A critical remote code execution (RCE) vulnerability chain in GitLab’s Jupyter Notebook diff renderer. This issue is rooted in two long-standing memory safety vulnerabilities within the Oj Ruby JSON parser. The vulnerabilities impact both GitLab Community Edition and Enterprise Edition…
Weekly Cybersecurity Newsletter – Top 50 Biggest Cybersecurity Stories– SonicWall Zero-Day, Cl0p Windchill Attack, AI-Weaponized Threats, Data Breaches & More
Welcome to this week’s edition of the GBHackers cybersecurity newsletter — your weekly cybersecurity bulletin covering the 50 most important stories from July 20–24, 2026. It was a heavy week: Cl0p turned internet-exposed Windchill servers into a global data-theft campaign,…
Hackers Use Stealer Logs to Bypass MFA and Launch Ransomware Attacks
Infostealer malware has now become the invisible thread linking petty credential theft to full-blown ransomware campaigns. Attackers no longer bother forcing their way through firewalls when infostealers have already unlocked the front door for them. Documented by DarkOwl, a stealer log…