Issues

• A second run() in one test drops the first run and its scores from the report

  getsentry/vitest-evals#108 · Filter by author marcusschiesser opened on Oct 7, 2026

• Option to write a compact or redacted eval report

  getsentry/vitest-evals#107 · Filter by author marcusschiesser opened on Oct 5, 2026

• Running multiple evals with different harness-configurations ("matrix run")

  getsentry/vitest-evals#101 · Filter by author JonasDoesThings opened on Sep 8, 2026

• Migrate deprecated @mariozechner/pi-* packages to @earendil-works/*

  getsentry/vitest-evals#88 · Filter by author ShravanSunder opened on Jul 13, 2026
  1

• Support AI SDK v7 (ai@7): widen peer range and make resolveOutput handle getter-backed results

  getsentry/vitest-evals#87 · Filter by author chentsulin opened on Jul 10, 2026

• Core framework gap analysis from agent-eval research

  getsentry/vitest-evals#86 · Filter by author dcramer opened on Jun 25, 2026
  4

• RFC: scenarioHarness — scripted and LLM-driven multi-turn evals

  getsentry/vitest-evals#82 · Filter by author tridha643 opened on Jun 10, 2026
  1

• Migrate build system from tsup to tsdown

  getsentry/vitest-evals#19 · Filter by author dcramer opened on Jul 12, 2025

• Add GitHub Action for JSON reporter

  getsentry/vitest-evals#13 · Filter by author dcramer opened on Apr 25, 2025
  5

• Support assertions on evals

  getsentry/vitest-evals#8 · Filter by author dcramer opened on Apr 22, 2025

• Add stateful eval reporter

  getsentry/vitest-evals#4 · Filter by author dcramer opened on Apr 13, 2025

• Native scorers

  getsentry/vitest-evals#2 · Filter by author dcramer opened on Apr 11, 2025