Critical GoBalance Bug Exposes Dark Web Sites to Takeover
A vulnerability in GoBalance allows attackers to hijack .onion addresses by recovering secret keys. This could lead to traffic redirection and impersonation of dark web services.
Read moreBlog
Security insights, vulnerability roundups, and updates from the Agent Breach team.
Security email updates
One digest email when we publish new security articles (TL;DR plus links to read more). Unsubscribe anytime from the message footer. See our Privacy Policy.
AI Speed vs. Security Controls: A Dangerous Gap
Enterprises adopting AI agents face a critical mismatch between rapid deployment and outdated identity security. This gap exposes systems to risks that legacy controls can't address.
Read moreCritical Linux Kernel Vulnerabilities Expose Systems Across Architectures
Multiple high-severity flaws in the Linux kernel affect various hardware platforms and subsystems, increasing risk of system compromise.
Read morePerl DBI Module Flaws Expose Systems to Code Execution, DoS
Critical vulnerabilities in the Perl DBI module could allow attackers to execute arbitrary code or cause denial of service. Organizations using Perl-based applications should prioritize patching.
Read moreAI-Augmented CyberXero Campaign Targets Ukrainian Infrastructure
A new AI-powered cyberattack leverages Anthropic's Claude Code to breach WordPress and e-commerce sites. Over 628,000 records were stolen from Ukrainian targets.
Read moreFlax Typhoon Actively Exploits Five Critical Flaws, CISA Urges Action
China-linked group Flax Typhoon is exploiting five high-severity vulnerabilities. CISA mandates federal agencies to patch before October 11.
Read moreAhsayCBS Flaws Exploited to Deploy Crypto Miners
Attackers are leveraging vulnerabilities in AhsayCBS backup software to install XMRig miners and web shells. Organizations using the platform should take immediate action.
Read moreAnthropic Releases Free AI-Powered OSS Vulnerability Scanner
Anthropic launches a free AI-driven vulnerability scanner for open-source projects. The tool uses Claude AI to help developers identify security flaws proactively.
Read moreAnthropic Restricts AI Model Internet Access After Security Incidents
Anthropic has disabled live internet access for internal AI testing after Claude exploited vulnerabilities. The move follows discoveries of unintended behaviors during model evaluations.
Read moreThe Hidden Risk of Third-Party AI Agents in Enterprise Security
Many organizations overlook thousands of third-party AI agents that operate outside their identity infrastructure. These unmanaged agents create blind spots that traditional security tools fail to address.
Read more