AgentHazard (Claude Code) - Internal Reconnaissance: leaderboard
Metric: Attack success rate (%) on the internal reconnaissance instances (using the agent to enumerate internal networks, ports and services), 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 | 24.51 |
| 2 | Kimi K2 | 24.51 |
| 3 | Qwen 2.5 Coder 32B Instruct | 68.63 |
| 4 | Qwen 3 VL 235B A22B Instruct | 72.55 |
| 5 | Kimi K2.5 | 84.31 |
| 6 | GLM-4.6 | 93.14 |
Interactive version: theaggregate.ai/benchmark?slug=agenthazard-claude-code-internal-reconnaissance · How It Works · Data refreshed daily, snapshot 2026-10-07.