Tag
The article describes testing whether an AI agent using CrewAI and RAG respects user instructions when a retrieved document contains conflicting commands, as part of efforts to develop security test cases.
The tweet introduces State Machines as a tool to solve the challenge of testing AI agents by providing stateful, repeatable enterprise environments for API mocking.
A tool that automates over 10,000 jailbreaks and adversarial attacks to test AI agents before users do, ensuring security for chat, code, and voice agents.
A developer built Battle Agents, a platform for testing AI agents in controlled failure scenarios to inspect decisions, tool calls, and recovery, and is seeking community feedback.