Agentjacking Attack Tricks AI Coding Agents Into Running Malicious Code

Cybersecurity researchers have identified a new attack vector called "agentjacking" that exploits AI coding assistants by tricking them into executing malicious code. The attack takes advantage of how these AI agents interpret and act on instructions, potentially allowing hackers to compromise systems through seemingly legitimate coding suggestions. This emerging threat highlights new security challenges as AI-powered development tools become more widely adopted across the industry.