README.md

metadata

title: AgentnessBench
emoji: 🎮
colorFrom: indigo
colorTo: purple
sdk: docker
app_port: 7860
pinned: false

AgentnessBench

A grid arena for measuring whether LLMs read other agents' motives. See docs/superpowers/specs/2026-06-01-proteus-arena-slice-design.md.

🎮 Live demo: https://irregular6612-agentnessbench.hf.space — play in the browser from anywhere (keyless public deploy: human play + offline persona/policy memory work; real-model spectate is disabled). Deploy notes: docs/DEPLOY-huggingface.md.

명령어 사용법 (한국어): docs/USAGE-ko.md — play / run / replay / compare 전체 안내.