Top AI Agents Built to Catch Malicious Code Can Be Tricked Into Running ItThe Hacker News·Jul 9, 05:17 UTC · Jul 9, 2026Exploit / PoCCVE-2026-3986150
Cursor Flaw Lets Malicious Cloned Repositories Trigger Windows Code ExecutionThe Hacker News·Jul 15, 00:00 UTC · Jul 15, 2026Exploit / PoCCVE-2026-26268CVE-2026-10591150
New Agent Data Injection Attack Can Make AI Agents Misclick or Run Attacker CommandsThe Hacker News·Jul 15, 00:00 UTC · Jul 15, 2026Exploit / PoCCVE-2025-32711150
Claude Code and Gemini CLI Flaws Let a GitHub Issue Reach CI Workflow SecretsThe Hacker News·Aug 7, 08:18 UTC · Aug 7, 2026Exploit / PoC in the wildCVE-2026-12537CVE-2026-54316160
More evidence your AI agents can be turned against youCyberScoop·Dec 5, 20:48 UTC · Dec 5, 2025Exploit / PoC in the wild60
Public GitHub Issue Could Trick GitHub Agentic Workflows Into Leaking Private Repo DataThe Hacker News·Jul 7, 14:07 UTC · Jul 7, 2026Exploit / PoC145