In Part 1 of this series, we introduced the AI agent as a High-Privilege Non-Human Identity (NHI). We argued that these agents are not just…

We ran a “clean” automated scan on a fintech client’s iOS and Android apps. Both came back green. Three days into the manual engagement, we…

Enumerate a running internal service, exploit a vulnerable web application, pivot through the system, and crack your way to root.

Slort is a Windows machine that chains a PHP remote file inclusion vulnerability with a world-writable scheduled task binary to deliver a…

A real finding from a penetration test on an enterprise integration platform

Learn mobile application pentesting through static analysis, MobSF, and the OWASP Mobile Top 10.

Apple updates its Hide My Email service by migrating addresses to a dedicated domain, making it easier for sites to identify private proxies.

A critical vulnerability in Monero P2Pool is actively exploited to steal mining rewards. Miners must urgently update to P2Pool v4.16 to secure their payouts.

China dismantled Silver Fox Trojan cells across five provinces, detaining dozens tied to phishing sites and a 439-domain CNCERT-tracked campaign.

Kaspersky warns of Steam Workshop malware disguised as Wallpaper Engine files, deploying DarkKomet RATs and info stealers to compromise systems.

A single link to a trusted Microsoft domain could quietly turn Copilot into a data exfiltration tool. Varonis Threat Labs disclosed this flaw, naming it SearchLeak. The chain let an attacker steal…

Aikido Security discovered 15 JetBrains malicious plugins stealing developer API keys. Learn how these fake AI assistants compromise your development environment.

A critical Google Vertex AI vulnerability in the Python SDK allowed attackers to hijack machine learning models. Discover how to patch this dangerous flaw.

A China-nexus threat group known as Velvet Ant hid inside one organization’s network for nearly a decade. Sygnia’s incident response team uncovered the campaign and named it Operation …

OpenAI introduces ChatGPT Scheduled Tasks, transforming the AI into a proactive agent capable of automated monitoring and routine management.

Microsoft's China AI business is thriving, with ByteDance spending $1 billion annually on OpenAI models via Azure despite US export bans.

A Moxa NPort vulnerability (CVE-2026-10829) lets authenticated attackers gain root via RCE on serial device servers. Patch v1.5.1 now.

As artificial intelligence dominates tech headlines, cybercriminals are riding the wave. FortiGuard Labs has detailed a fresh campaign built around AsyncRAT AI lures — booby-trapped archives dress…

Summary

A web server shipped its .git folder by accident — and Git never forgets.