When Your AI Agent Goes Rogue: The Hidden Risk of Excessive Agency

In Oct 2025, a malicious code in AI agent server stole thousands of emails with just one line of code. The package, called postmark-mcp, looked completely legitimate. It worked perfectly for 15 versions. Then, on version 1.0.16, the developer slipped in a tiny change. every outgoing email now included a hidden BCC to an attacker-controlled address. By the time anyone noticed, roughly 300 organizations had been compromised. Password resets, invoices, customer data, internal correspondence.

How to Align Your DevSecOps Framework with Software Supply Chain Security

A strong DevSecOps framework integrates security into every stage of the software development lifecycle (SDLC). But as development accelerates, reliance on third-party and open-source code grows, introducing significant risks from the software supply chain. Aligning your DevSecOps framework to address these specific threats is no longer optional. It’s essential for building resilient and secure applications.

10 AI SOC Benefits That Actually Transform Security Operations

See how Torq harnesses AI in your SOC to detect, prioritize, and respond to threats faster. Request a Demo Your SOC is drowning. Industry estimates suggest that up to 60% of SOC analyst time is spent on Tier 1 triage, leaving less time for addressing real threats. According to Splunk’s State of Security 2025 report, 59% of security teams report being overwhelmed by too many alerts, and 55% waste precious hours chasing false positives.

Uncovering A Mass VPN Phishing Campaign - The 443 Podcast - Episode 355

This week on the podcast, we cover some first-hand research from the WatchGuard Threat Lab on a phishing campaign targeting users of nearly every major VPN vendor. After that, we discuss two recently resolved vulnerabilities in the Fortinet FortiSIEM application, then end with research from Varonis on a new attack flow against Copilot called RePrompt.

Facebook tech support scams on the rise: How cybercriminals are turning your feed into a trap

From fake ads to tech support fraud, see how scammers exploit Facebook and how to protect your data and money. Facebook may feel like a safe place to connect, but scammers are increasingly using its ads, posts, and messages to deceive users. Here’s how cybercriminals are turning your feed into a gateway for fraud and what you can do to stay protected. When you open Facebook, you might expect birthday alerts, travel snapshots, or quick messages with friends.

A New Security-Focused Path Search View for Faster, Clearer Analysis

Security teams frequently investigate whether traffic is allowed, denied, or translated along a path, but traditional visualization methods create unnecessary complexity. A standard Path Search view includes every L2 and L3 hop—switches, routers, and intermediary devices that participate in forwarding but apply no security policy. While this depth is helpful for network troubleshooting, it overwhelms analysts focused specifically on enforcement points.

Switch from SonicWall and Succeed

If you use SonicWall, you know what type of year 2025 was for your business. Between breaches, price increases, and confusing pivots, your vendor left you to clean up their mess while you paid more for their product and earned less in return. Join Adam Winston, Field CTO at WatchGuard Technologies, to learn why our Unified Security Platform is the right choice for a profitable, worry-free 2026. In this session, we’ll discuss.

PCI DSS Penetration Testing Requirements Explained

Overall, PCI DSS 4.0.1 is a set of 12 requirements distributed over six goals as a security standard for credit cards and debit cards. Not having proper documentation, poor protocols, or insufficient penetration testing may be among the reasons as to why PCI DSS audits fail.

Custom Risk Scoring Is the Missing Link Between Disconnected Findings and Real Exposure Management

Most large organizations rely on multiple vulnerability and exposure scanning tools out of necessity. Infrastructure scanners, cloud security platforms, application security testing tools, container scanners, and attack surface management solutions all play a role. Each one is designed to answer a specific question. But when it comes to understanding the risk of the vulnerabilities and exposures they detect, each tool has its own approach to quantifying it.

Apono + MongoDB: Secure Access Across MongoDB, Atlas, and Atlas Portal

MongoDB powers some of the world’s most modern applications.Everything from self-managed deployments to fully managed cloud environments run with MongoDB Atlas. But as teams scale across environments and projects, managing secure access becomes increasingly complex. Apono brings Just-in-Time, least-privilege access to MongoDB services across MongoDB, MongoDB Atlas, and MongoDB Atlas Portal.