Machine solving traces and results generated by BoxPwnr.
Each trace includes the full conversation log showing LLM reasoning, commands executed, and outputs received. Browse leaderboards, replay runs in an interactive web viewer, and read AI-generated reports:
| Platform | Solved | Completion | Traces |
|---|---|---|---|
| HTB Labs | 54/514 |  |
326 |
| HTB Starting Point | 25/25 |  |
772 |
| PortSwigger Labs | 163/270 |  |
377 |
| XBOW Validation Benchmarks | 94/104 |  |
512 |
| Cybench CTF Challenges | 35/40 |  |
616 |
| picoCTF Challenges | 115/439 |  |
428 |
| TryHackMe Rooms | 31/459 |  |
238 |
| HackBench Benchmarks | 3/16 |  |
3 |
| Neurogrid CTF: The ultimate AI security showdown | 17/36 |  |
197 |
Last updated: 2026-02-23 22:10:05