Trial Benchmark Dual-use
Cybench
Cybench is a benchmark and agent framework from Stanford and collaborators for evaluating model capabilities on 40 CTF-style cybersecurity tasks, with optional subtask-level scoring. It is used by researchers and AI labs to measure cyber-offense risk of LLM agents.
Worth piloting on a real project; expect some rough edges.
Facts
- Repository
- andyzorigin/cybench
- Homepage
- github.com
- Listed via
- Curated
- First seen
- 2026-11-08
- License
- Apache-2.0
- Stars
- 355
- Last commit
- 2026-10-17
- Checked
- 2026-11-10
Listing is not endorsement. Summaries and notes are AI-assisted and reviewed; repository facts refresh automatically.