Google Makes CodeMender Available as Managed AI Security Agent

Google's CodeMender, developed by DeepMind, has transitioned from a research project into a fully managed, enterprise-ready AI code security agent available in public preview on the Gemini Enterprise Agent Platform. The agent combines static and dynamic analysis with Gemini models to discover, validate (including running PoC exploits in customer sandboxes), and automatically generate tested patches; it integrates with Google Cloud, Mandiant and Wiz, supports major languages and frameworks, offers secure deployment controls and a restricted pilot security-tuned model, and plans expanded governance and orchestration features.