A clinical decision-evidence benchmark: pull the one fact an answer rests on, and see whether the assistant's action moves with the evidence. Built on HealthBench.
xinxuxin/keystone-bench has 499 GitHub stars. It gained +92 stars over 6 recorded days in the last 7 days (snapshots through Oct 3, 2026).
Its biggest recent day was Oct 1, 2026, with +21 stars.
It currently ranks #47491 daily, #47428 weekly, #47837 monthly by momentum on Trending Repos; its best-ever rank is #924.
Trending Repos first recorded it on Sep 27, 2026.
| Date | Stars gained | Total stars | Forks gained |
|---|---|---|---|
| +19 | 494 | 0 | |
| +19 | 475 | 0 | |
| +21 | 456 | 0 | |
| +13 | 435 | 0 | |
| +20 | 422 | 0 | |
| 0 | 402 | 0 |