A3S-Bench

What is A3S-Bench?

A3S-Bench is an executable security test suite for evaluating autonomous agents across multi-turn scenarios and agent-level risks using an action-grounded…

Released

2026-05-21

Evaluates

Agents, Agents & Tool Use, General AI

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2605.22321

A3S-Bench paper and code

• Paper https://arxiv.org/abs/2605.22321
• Code https://github.com/antgroup/Agent3Sigma-Stage

Which sources cite A3S-Bench?

• A3S-Bench

Related Agents, Agents & Tool Use, General AI benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• AARRI-Bench
• ActBench
• ActiveSciBench
• ActiveVision
• AD-Diff Bench