AutoPenBench

Evals

Benchmark / dataset

33 pentest tasks for generative agents, in-vitro and real CVEs.

Category
Cyber offense capability
Insurance exposure
Breach/cyber
Publisher
Politecnico di Torino
Publisher type
Academic
License
MIT
Status
Maintained (<12mo)
Last commit
2025-10-28
Stars
104

View source on GitHub →

Wanich Risk Lab, 2026.