This website uses cookies
Read our Privacy policy and Terms of use for more information.
All of our previous posts
Nov 15, 2026
•
6 min read
Four AI attack frameworks hit real victims in one week, banks among them, and each one got in through something left open and stalled at a working login.
Nov 8, 2026
Microsoft watched an AI agent go from a leaked GitHub secret to an emptied Azure Key Vault in five minutes, while a dozen AI platforms lost customer tokens that attackers reused within half an hour.
Nov 1, 2026
A tampered Claude Code extension reached up to 5,000 Cursor users in eleven hours and a planted AGENTS.md hijacked coding agents, showing the developer laptop is now the shortest path to production.
Oct 25, 2026
A LiteLLM master key sat unrevoked for 136 days before attackers used it to ship a malicious release, and ConnectWise shows what AI agents do once they hold a stolen credential.
Oct 18, 2026
Anthropic, OpenAI and CERT-In disclosed AI agents breaching real organisations during testing, mostly through keys left in public rather than zero-days, and the victims' SOCs saw none of it.
SEC://
Oct 13, 2026
8 min read
Cyber Security
+2
Oct 11, 2026
LLM proxies, AI gateways, agent runtimes and MCP servers got hit in the wild this week: 3,000+ servers infected, a CVSS 9.9 and one prompt for IAM keys.
Oct 4, 2026
An AI agent breached a vulnerability-disclosure nonprofit in seconds, an AI-found bug was exploited within a day of disclosure, and your monthly patch cycle never had a chance.
Sep 27, 2026
A criminal spent $12,000 on three open-source AI agents to breach 27+ companies and steal 600,000 cards — plus why the OpenAI "agents attacking the internet" panic is aimed at the wrong story.