Skip to content

fix: auto-derive evaluator name from type when not provided#969

Merged
christso merged 2 commits intomainfrom
fix/968-auto-name-evaluators
Apr 8, 2026
Merged

fix: auto-derive evaluator name from type when not provided#969
christso merged 2 commits intomainfrom
fix/968-auto-name-evaluators

Conversation

@christso
Copy link
Copy Markdown
Collaborator

@christso christso commented Apr 8, 2026

Summary

Test plan

  • All 1410 core tests pass
  • E2e dry-run with unnamed evaluators (latency, contains, rubrics) — no "Skipping evaluator with missing name" warnings
  • Pre-push hooks pass (build, typecheck, lint, test)

🤖 Generated with Claude Code

christso and others added 2 commits April 8, 2026 04:37
Previously, evaluators without an explicit `name` field were skipped with
a warning for non-assertion types (llm-grader, code-grader, latency, etc.).
Now `generateAssertionName` falls back to the type name itself, so all
evaluator types work without requiring a name.

Also cleans up test warnings: migrates deprecated `required_min_score` to
`min_score` and suppresses intentional skip-behavior console output.

Closes #968

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@christso christso merged commit 33bd3ce into main Apr 8, 2026
3 of 4 checks passed
@christso christso deleted the fix/968-auto-name-evaluators branch April 8, 2026 04:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Evaluator name should be optional — auto-derive from type + position/value

1 participant