Red teaming & safety
Probe the agent with adversarial inputs; layered guardrails block them and ASR measures what slips through.
Section: testing-evaluation · scene id red-teaming-safety · tutorial 04-testing-evaluation/04-red-teaming-safety
Probe the agent with adversarial inputs; layered guardrails block them and ASR measures what slips through.
Section: testing-evaluation · scene id red-teaming-safety · tutorial 04-testing-evaluation/04-red-teaming-safety