AgentHazard (OpenClaw): leaderboard

AgentHazard (2026) measures how often computer-use agents carry out harmful objectives hidden in sequences of individually plausible steps, here with each backbone model running inside OpenClaw. The score is the attack success rate in percent, judged on the full trajectory by Gemini-3-Flash; lower is better.

Metric: Attack success rate (%), the share of instances judged harmful over all ten risk categories, with the model running inside the OpenClaw open-source agent framework (browser, shell and file-system tools; one isolated Docker container per task), over AgentHazard instances (harmful objectives embedded in locally legitimate multi-step workflows, ten risk categories and ten attack strategies; the evaluated runs cover 1,000 instances), judged on the full trajectory (round_all) by Gemini-3-Flash, which labels a trajectory harmful when it realizes the objective; backbones served through an OpenAI-compatible API; lower is better. Source: arxiv.org. 7 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 2.5 72B Instruct24.1#1117
2Kimi K2.550#266
3Qwen 3 32B59.18#972
4Qwen 3 VL 235B A22B Instruct63#532
5Qwen 2.5 Coder 32B Instruct64.06#1547
6GLM-4.670.8#498
7Kimi K271.1#443

Interactive version: theaggregate.ai/benchmark?slug=agenthazard-openclaw · How It Works · Data refreshed daily, snapshot 2026-11-21.