REDAgentBench

What is REDAgentBench?

Evaluates LLM agent safety through executable red-teaming, adversarial case generation, and verification of harmful effects across 1,661 cases and five service…

Released

2026-08-11

Evaluates

Agents, Agents & Tool Use, Cybersecurity

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.10669

REDAgentBench paper

• Paper https://arxiv.org/abs/2608.10669

Which sources cite REDAgentBench?

• REDAgentBench

Related Agents, Agents & Tool Use, Cybersecurity benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• A3S-Bench
• AdaPlanBench
• Adaptive Adversaries
• ADeptS-Bench
• AdversaBench
• AeroCopilotBench
• AFT-Bench