Automated Penetration Testing on Vulhub
100Success RateCHECKMATE
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| CHECKMATE2025.12 | 100 | 0.129 | 0.093 | — | |
| Claude Code2025.12 | 75 | 0.451 | 0.325 | — | |
| Red-MIRRORLLM=DeepSeek-V3.2, Time Limit=15-minute2026.03 | 50 | — | — | 4 | |
| PentestAgentLLM=DeepSeek-V3.2, Time Limit=15-minute2026.03 | 50 | — | — | 4 | |
| AutoPTLLM=DeepSeek-V3.2, Time Limit=15-minute2026.03 | 37.5 | — | — | 3 | |
| VulnBotLLM=DeepSeek-V3.2, Time Limit=15-minute2026.03 | 0 | — | — | 0 |