OpenEval 0.3.2

Fixes collapsed session tool-group labels with the production OpenCode 2.0.3 UI translations. Uses the upstream production grouping names and plural count, with source provenance recorded for the vendored session UI. Includes the OpenCode 2.0.3 upgrade from 0.3.1. Verified retained and production session rendering, code-judgment inspection, and desktop/mobile layouts.

OpenEval 0.3.1

Pins the SDK, client, schema, server, UI, and related OpenCode dependencies to production 2.0.3. New candidate runs use openeval-runtime:2.0.3; rebuild it with the image command after upgrading. Native archive readers keep recorded version metadata and let the SDK open supported schemas on disposable copies. Scoped benchmark completion now respects admitted work, allowing retained results from older runtimes to remain selected.

OpenEval 0.3.0

Code and hybrid judging is now supported. Export an ordinary function from judge.ts, use judge.md for LLM criteria, or use both against the same recorded EvalRun. Boolean and fractional criterion scores use equal-weight aggregation. JudgeContext exposes recorded metrics, traces, files, full native history, SDK/schema access, and read-only SQL. The viewer now shows code output, frozen source, and process logs. This pre-1.0 release uses canonical criterion names and results schema 5; older stores require a separate migration.