Stories cluster related articles into a single narrative, linked to the underlying CVEs and affected products. 8,564 stories synthesized.
Check Point Research's latest bulletin highlights a range of cyber incidents, including attacks on airports, government agencies, and medical device companies, alongside new AI-driven threats and critical software vulnerabilities.
Slovenia's largest gambling and tourism group, Hit, has begun reopening its casinos after a cyberattack forced a multi-day shutdown of critical gaming systems and other IT infrastructure.
Amazon Web Services' new Console Private Access feature, designed for secure access within isolated VPCs, may inadvertently prevent users from signing into their personal AWS accounts if not configured properly.
Cybercriminal group ShinyHunters claims to have exfiltrated 284 million patient records from major US healthcare distributor McKesson, following a breach detected on August 25, 2026.
A supply-chain attack has compromised the @7nohe/openapi-react-query-codegen npm package, injecting a credential-stealing worm dubbed Trinitite that targets developer workstations and build systems.
Anthropic's new Compliance API for Claude Code provides security teams with crucial visibility into local AI agent actions, though it highlights challenges in verifying legitimate access.
A critical arbitrary file read vulnerability in Ruby on Rails, dubbed KindaRails2Shell and tracked as CVE-2026-66066, is being actively exploited by attackers to steal secrets and achieve remote code execution.
A sophisticated Magecart campaign, dubbed HexMage, is using the Ethereum blockchain to host malicious JavaScript loaders, enabling the theft of credit card data from over 40 online merchant websites globally.
A novel prompt injection technique embeds malicious AI instructions within the text of legal documents, posing a new threat to AI system integrity.
OpenClaw 2.0, the latest release of the open-source AI agent platform, introduces significant security enhancements for AI agents, plugins, and credential management.
Attackers are distributing the sophisticated ValleyRAT backdoor by disguising its installer as legitimate adware, employing DLL sideloading and disabling Windows Defender to achieve execution.
Russian state-sponsored hackers are employing a novel technique, dubbed GuardBreaker, to circumvent AI-powered malware analysis tools by embedding manipulative prompts within code comments.
Key findings • 17 malicious npm packages were disclosed in under one minute. • All packages were assigned a Critical severity rating. • Packages were recently published, indicating fresh …
A new Windows malware toolkit, Gryxa, leverages AI to steal credentials, maintain persistence, and actively observe and counter security team remediation efforts.
Microsoft has confirmed a widespread bug causing Microsoft Defender Antivirus to falsely report that protection is turned off, despite the software functioning correctly.
A China-nexus threat actor known as Fire Ant has broadened its cyber espionage operations to target Cisco IOS XR routers and TACACS servers, aiming to steal credentials and disable security logging.
August 2026 saw significant cybersecurity events including a data breach at Hugging Face, ongoing threats to critical infrastructure, and a novel attack using spoofed in-flight Wi-Fi.
The Rhysida ransomware group claims to have stolen over 5TB of sensitive data from an unnamed victim, who has publicly stated they will not pay the extortion demand.
A flaw in the PHP dependency manager Composer, CVE-2026-59944, enables malicious packages to alter file permissions, potentially exposing sensitive data like SSH keys on shared systems.
A critical vulnerability (CVE-2026-73296) in Microsoft's UFO automation framework allows unauthenticated attackers to remotely control Android devices via exposed MCP services.
Key findings • Three vulnerabilities disclosed on August 31, 2026, affect Code Projects' Simple Inventory System 1.0 and Employee Leave Managing System 1.0. • Two Cross-Site Scripting (XSS) v…
A recent swarm attack targeting Hugging Face's AI models serves as a stark warning for the evolving landscape of AI security, demonstrating the potential for sophisticated, autonomous threats.
Dr. Joye Purser of Cohesity details a pragmatic approach to vulnerability management, balancing KEV, EPSS, and CVSS scores with real-world risk factors.
A new open-source Python package, halo-record, provides immutable, cryptographically hashed audit trails for AI agents to enhance transparency and accountability.