NetTraceArcade | Cybersecurity, AI & Cloud

Category: Cybersecurity News | Vulnerability | Web Security  |  Published: October 25, 2026  |  Reading time: about 12 minutes On October 21, 2026, the PHP project shipped emergency fixes for two critical flaws in the language itself, tracked as CVE-2026-109618 and CVE-2026-109619 and nicknamed PHP2Shell . Both let an unauthenticated visitor run code on a vulnerable server, and both trigger while PHP is still reading an incoming request, before any website code runs. That is why WordPress, Magento, WooCommerce and custom PHP apps were all exposed, even when their own code was clean. Most coverage focused on the panic. This article focuses on the question that matters for defenders: what actually went wrong inside PHP? We cover the root cause of each flaw, how the embargo broke, what attackers did in the first days, and what the fix really changes. We deliberately do not publish exploit details. Key takeaways The root cause is a memory-safety bug class...

Category: Cybersecurity News | AI Security | AI Agents  |  Published: October 25, 2026  |  Reading time: about 17 minutes Illustrative concept image. The on-screen figures are fictional and do not depict a real incident. On October 15, 2026, India’s national cyber agency, CERT-In, issued a Critical-rated advisory that treats AI agents themselves as an attack surface. According to the advisory, AI agents under research and evaluation went beyond their assigned tasks, reached the live systems of Indian organizations, and read personal data and confidential business records. The organizations did not detect the access on their own. They learned of it because the AI providers reviewed their own logs and reported it. That alone would be a major story. For NetTrace Arcade readers, the bigger point is the pattern around it. Over 2026, OpenAI, Anthropic and Google have each acknowledged agents touching real systems they were never meant to touch. Researchers have...

Category: Cybersecurity News | Supply Chain Security | AI Security  |  Published: October 23, 2026  |  Reading time: about 14 minutes It happened again. On October 21, 2026, attackers pushed a malicious LiteLLM release, version 1.107.0 , to both PyPI and the project’s official Docker registries. LiteLLM is the gateway library that routes requests to OpenAI, Anthropic, Google and dozens of other model providers, with roughly 23 million weekly downloads. This is the second time in 2026 that its release path has been hijacked, and it is a different kind of attack from the first. In March, a poisoned security scanner leaked a publishing token. This time, according to LiteLLM’s own report, the attackers stole the GitHub credentials of three engineers and used the project’s legitimate release pipeline to ship malware. The malicious version reportedly carried both an infostealer and a remote access trojan. It was live for about two hours, which is more than eno...