Tenet’s ‘Agentjacking’ Attack Turns Sentry Errors Into Code Execution

AI coding agents can create a new code execution risk when treating externally influenced error data as trusted guidance. This 'Agentjacking' attack can be used to inject malicious code. It's a concern when agents have access to command line tools. Developers should be cautious when using AI coding agents and ensure they don't blindly trust error data.