Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

We found a path traversal vulnerability in Smithery.ai that compromised over 3,000 MCP servers and exposed thousands of API keys. Here's how a single Docker build bug nearly triggered one of the largest AI supply chain attacks to date.

Social media is so deeply intertwined into our lives that as our online presence grows, so does the opportunity for cybercriminals to exploit it. Our data shows just how widespread social media hacking is and who hackers typically target.

I’ve spent my career building and defending. I’ve seen the beauty of innovation and the brutality of cyber warfare. I’ve sat shoulder-to-shoulder with security engineers and platform teams in the trenches at 3AM responding to state-backed attacks where context and speed meant survival. In those moments, one truth becomes painfully clear: With AI, that gap just became a chasm. Agentic AI didn’t bend the rules of software — it rewrote them. Code now evolves in real time.

According to a new Ponemon study, weak file protections now account for several cybersecurity incidents a year for many organizations. Unsafe file-sharing practices, malicious vendor files, weak access controls, and obscured file activity are largely to blame. File Integrity Monitoring (FIM) could be the solution.

We’re excited to announce that Gartner has named JFrog a ‘Visionary’ in the 2025 Magic QuadrantTM for Application Security Testing. We believe this reflects JFrog’s forward thinking strategy of integrating application security seamlessly throughout the entire software development lifecycle in ways that help organizations deliver their most secure, trusted applications without impacting developers’ productivity.

In its 2025 State of the Underground report, Bitsight TRACE found that ransomware activity continued to escalate in 2024, with a 25% increase in unique victims listed on leak sites and a 53% increase in the number of ransomware group-operated leak sites. The report also observed a 43% increase in data breaches shared on underground forums, with nearly one in five victims based in the United States. These findings highlight a continued upward trend in cyberattack activity.

Astaroth trojan uses GitHub to host malware configurations, TA585 delivers MonsterV2 malware in phishing campaigns, and threat actors exploit Microsoft’s logo in tech support scams.

Threat actors are abusing X’s generative AI bot Grok to spread phishing links, according to researchers at ESET. The attackers achieve this by tricking Grok into thinking it’s answering a question, and providing a link in its answer. “In this attack campaign, threat actors circumvent X’s ban on links in promoted posts (designed to fight malvertising) by running video card posts featuring clickbait videos,” ESET says.