Scenario: Open-source framework for automated AI app red-teaming

LangWatch released Scenario, an open-source framework that automates multi-turn red-team exercises against production AI agents. The tool uses a four-phase "Crescendo" escalation and persistent attacker memory to simulate realistic social-engineering-style attacks across conversational turns, aiming to find compromised agents with access to databases or financial tools; it targets banks, insurers, and AI-first companies and is designed to integrate into CI pipelines.