AgentHazard (Claude Code) - Privilege Escalation: leaderboard
Metric: Attack success rate (%) on the privilege escalation instances (escaping containment or obtaining elevated privileges), with the model running inside the Claude Code CLI agent (persistent shell, file editing, tool use; restricted shell sessions), 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.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 19.59 |
| 2 | Kimi K2 | 28.87 |
| 3 | Qwen 3 VL 235B A22B Instruct | 55.67 |
| 4 | Qwen 2.5 Coder 32B Instruct | 62.89 |
| 5 | GLM-4.6 | 77.32 |
| 6 | Kimi K2.5 | 79.38 |
Interactive version: theaggregate.ai/benchmark?slug=agenthazard-claude-code-privilege-escalation · How It Works · Data refreshed daily, snapshot 2026-10-07.