Post

Cybersecurity Newsfeed - 11/05/26

Daily cybersecurity news covering vulnerabilities, adversaries, trends, breaches, and other notable security developments.

Cybersecurity Newsfeed - 11/05/26

Cybersecurity Newsfeed

📅 11/05/26

🛡️ Vulnerabilities

  • Ollama Out-of-Bounds Read: A high-severity flaw was discovered in Ollama, a framework for running LLMs locally. The bug involves improper handling of input tensors during inference, which could lead to system crashes or memory leaks. More info

  • “Dirty Frag” Linux Kernel Exploit: Qualys researchers detailed a technique that exploits the Linux kernel’s page cache. By manipulating memory fragmentation, attackers can achieve unauthorized data access or privilege escalation. More info

  • CISA KEV Update (CVE-2026-42208): CISA added a new vulnerability to its Known Exploited Vulnerabilities catalog. This flaw allows for remote code execution and is being actively leveraged by advanced persistent threat actors. More info

  • Edge Plaintext Password Storage: Security researchers found that Microsoft Edge stores passwords in plaintext within system memory. Microsoft stated this is “by design,” leading to debates regarding defense-in-depth strategies. More info

  • Analysis of CVE-2025-68670: Kaspersky provided a deep dive into this vulnerability, which is used in targeted attacks to bypass Windows security mitigations and achieve kernel-level execution. More info

🎯 Adversaries

  • CrateStealer Targets macOS via Google Ads: Cybercriminals are using fraudulent Google Ads for “Claude AI” to lure macOS users to phishing domains that deliver Python-based CrateStealer malware. More info

  • Fake OpenAI Repository on Hugging Face: A typosquatted repository impersonating OpenAI was found distributing infostealers. It targets AI developers to exfiltrate session tokens and environment variables. More info

  • RansomHouse Claims Trellix Breach: The extortion group claims to have breached cybersecurity firm Trellix, sharing screenshots of internal systems as proof of data exfiltration. More info

  • TCLBanker Hits Financial Institutions: A new banking trojan is using sophisticated redirection and script injection to capture multi-factor authentication codes and banking credentials in real-time. More info

  • Pamdoora Linux Backdoor: This new malware utilizes Pluggable Authentication Modules (PAM) to maintain stealthy persistence and intercept login credentials on Linux servers. More info

  • ClickFix “Troubleshooting” Scams: A new social engineering tactic uses fake macOS troubleshooting sites to trick users into running malicious terminal commands that steal iCloud session tokens. More info

  • AI Supply Chain Risks: A security incident at Braintrust has raised significant alarms regarding the integrity of the AI software supply chain and the potential for downstream client impact. More info

  • Abuse of Software Distribution Hubs: The hijacking of the JDownloader site to bundle malware with legitimate installers highlights the growing threat of web-based supply-chain attacks. More info

💥 Breaches & Leaks

  • Zara Data Breach: Fashion retailer Zara disclosed a breach affecting 197,000 individuals. The exposure originated from a third-party marketing database and included contact details and transaction history. More info

  • NVIDIA GeForce NOW Exposure: NVIDIA confirmed a localized data breach affecting users in Armenia due to a regional service provider’s infrastructure compromise. More info

  • ClickUp API Key Leak: Improperly secured environment files and public repositories led to the exposure of ClickUp API keys, potentially allowing unauthorized access to private project data. More info

📚 Others

  • CrimeNetwork Marketplace Dismantled: International law enforcement successfully shut down a rebooted version of the CrimeNetwork underground market and arrested its primary administrator. More info

⬅ Back to Archive

This post is licensed under CC BY 4.0 by the author.