Securing the AI Agent: A Unified Framework for Multi-Layer Agent Red Teaming Paper • 2606.31227 • Published Jun 30 • 14