[Experiment] Dart evals#22

Draft

ericwindmill wants to merge 13 commits into

mainfrom

ew-dart-evals-experiment

Commits on May 12, 2026

docs: update framework documentation to reflect auto-resolved AI backends, simplified API, and modular package structure
ericwindmill
committed

Commits on May 13, 2026

feat: add sandbox workspace persistence to eval runs via EvalConfig
ericwindmill
committed

Commits on May 14, 2026

feat: support multi-provider models in EvalSet and add comprehensive example evaluation suite
ericwindmill
committed
feat: add comprehensive evaluation logs and refactor sandbox/evaluation framework components
ericwindmill
committed
refactor: enable incremental result processing via onResult callback to write trajectories and logs per cell
ericwindmill
committed