AI coding agents' 0-click RCE flaw enables attacker control

This vulnerability allows attackers to remotely execute code in AI coding agents, posing significant security risks.

Auto-scored: The flaw represents a real escalation in security risk with no mitigations reported.