ACES (Agent Capability Evaluation Suite)

Evals

Framework / harness (not an eval — tooling/reference entry)

Harness to benchmark and RL-train security agents; recommended runner for ExCyTIn-Bench.

Category
Cyber defense & threat intel
Insurance exposure
Breach/cyber
Publisher
Microsoft
Publisher type
Big tech
License
MIT
Status
Active (<90d)
Last commit
2026-09-15
Stars
18
Notes
Released Aug 2026.

View source on GitHub →

Wanich Risk Lab, 2026.