What's Changed
Features
- feat(experiments): rebuild the results grid around the score columns by @nkabardin in #16917
- feat(experiments): a worse-than-comparison filter and a score-by-run matrix by @nkabardin in #16918
Fixes / Improvements
Chores
- chore(deps): bump uuid to 14.0.2 by @nimarb in #17086
- docs(api): advertise self-hosted API reference by @wochinge in #17084
- docs(agents): rank by the handbook, and stop inventing staleness by @nkabardin in #17075
- docs(agents): make verification proportionate to risk by @nkabardin in #17064
Full Changelog: v4.29.0...v4.30.0