AgentHazard (Claude Code) - Persistence Establishment: leaderboard

AgentHazard attack success rate on persistence establishment attempts (backdoors, SSH keys, startup scripts or scheduled jobs that keep long-term access), with each backbone model running inside Claude Code. Percent of instances judged harmful on the full trajectory by Gemini-3-Flash, lower is better.

Metric: Attack success rate (%) on the persistence establishment instances (backdoors, SSH keys, startup scripts or scheduled jobs that keep long-term access), 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. 6 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Qwen 2.5 72B Instruct22.22#1117
2Kimi K230.3#443
3Qwen 2.5 Coder 32B Instruct59.6#1547
4Qwen 3 VL 235B A22B Instruct68.69#532
5Kimi K2.580.81#266
6GLM-4.689.9#498

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