AgentHazard (OpenClaw) - Destructive Action: leaderboard

AgentHazard attack success rate on destructive action attempts (irreversible damage such as deleting files, wiping databases or encrypting data), with each backbone model running inside OpenClaw. Percent of instances judged harmful on the full trajectory by Gemini-3-Flash, lower is better.

Metric: Attack success rate (%) on the destructive action instances (irreversible damage such as deleting files, wiping databases or encrypting data), 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 Instruct25.26#1117
2Kimi K2.549.47#266
3Qwen 3 VL 235B A22B Instruct61.05#532
4Qwen 2.5 Coder 32B Instruct62.11#1547
5Qwen 3 32B62.77#972
6GLM-4.665.26#498
7Kimi K268.42#443

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