fathom-lab/styxx

Instruments that check what AI agents claim against committed bytes. The diff gate reads a PR summary against its diff and reports by default. EXTERNAL-1 upheld 23 of 100 sampled accusations, precision 0.23 (85 path claims, styxx 7.47.0; path accusations withheld since). Receipts, failures included, are in the repo.

2 best points · 0 comments · 1 Hacker News post · 15 GitHub stars · View on GitHub

Hacker News discussions

• Styxx – text-only drift detector 0.916 AUC beats hidden-state baseline (2 points, 0 comments, by fathomlab, 2026-04-24)