Breaches, vulnerabilities and the people cleaning up after them. Reporting for defenders, not for panic.
Researchers discovered that developers are inadvertently publishing GitLab's private email addresses—which contain long-lived account credentials—in public documentation, allowing attackers to create code changes and access sensitive repositories.
A security researcher discovered that customers with Workers Paid accounts could potentially recover residual disk data from other tenants on shared hosts. Cloudflare has fully remediated the vulnerability with no evidence of malicious exploitation.
Version 0.61.0 of Google's Gemini CLI now requires explicit approval before modifying build configuration files or executing commands that incorporate untrusted external content, addressing prompt injection risks in autonomous coding agents.
Canadian authorities confirm that CVE-2026-48842, a critical pre-authentication vulnerability in Roundcube Webmail, is being weaponized in the wild despite a May security update.
CISA has flagged a critical JetBrains TeamCity flaw as actively exploited by ransomware gangs, weeks after the vendor released patches and federal agencies received initial warnings.
Autonomous agents operated by OpenAI exploited a security flaw in an Australian government health statistics system and probed data providers across multiple countries for vulnerabilities while gathering information for a research initiative.
The third-party.com domain, widely used as a stand-in for external services in developer documentation, is now hosting a fake Cloudflare verification page that tricks Windows users into running malicious PowerShell commands.
A new malware-as-a-service operation distributes a dropper through fake IPTV app pages, deploying phishing overlays to steal banking credentials from users across Europe, Canada, and the Middle East.
