feat(api): update API spec from langfuse/langfuse f0ea0c3

[langfuse-bot]

Generated from langfuse/langfuse prep-js-v6-py-v5-sdk-majors at f0ea0c3.

Confidence Score: 4/5

Update the stale score response test before merging so CI can collect and run it.

Summary

This PR changes the generated score API to accept lists only.

• Score creation now uses one list and returns batch results.

_{Reviews (1) · Last reviewed commit: "feat(api): update API spec from langfuse..." · Reviewed by Greptile}

[github-actions]

@claude review

[claude]

Claude Code Review

No review was started: this request came from a bot account. Manual reviews can only be requested by someone with write access to this repository. Ask a maintainer to comment @claude review, or have your automation post the comment from a user account with write access.

_{Tip: disable this comment in your organization's Code Review settings.}

[claude]

Apart from the inline finding, I checked the one production caller of scores.create(), _upload_batch in langfuse/_task_manager/score_consumer.py. It isn't affected: it already sends a list cast to CreateScoresRequest, and it only branches on CreateScoreBatchResults, which is still in the response union.

Extended reasoning...

score_consumer.py doesn't import either removed type. Its 207 partial-failure handling relies only on CreateScoreBatchResults, which this diff keeps.

Additional findings (outside the current diff — GitHub can't attach inline comments there):

• 🔴 tests/unit/test_scores_batch_response.py — After merging, the unit test run fails: this test module cannot be imported, so none of its scores-create parsing tests run. Line 22 imports langfuse.api.scores.types.create_score_response, and this PR deletes that module. The single-200 case at line 31 and the CreateScoreResponse branch in _assert_parsed at lines 55-57 also test single-score input and the 200-with-id response, which the new contract drops. Fix: update the test to the list-only contract. Remove the import, SINGLE_REQUEST, the single-200 param and the status-200 branch, and keep the batch-202 and batch-207 cases.

  Why this was flagged

  tests/unit/test_scores_batch_response.py:22 runs from langfuse.api.scores.types.create_score_response import CreateScoreResponse. This diff deletes langfuse/api/scores/types/create_score_response.py and removes CreateScoreResponse from every __init__.py export. pytest therefore hits ModuleNotFoundError while collecting the module, and the whole file errors. That includes the sync and async batch-202 and batch-207 tests, which are the only checks on how the CreateScoresResponse union parses, and they no longer run. On the base branch this module imports fine and all six parametrized cases pass. Even with the import fixed, the single-200 case (line 31) sends a bare CreateScoreRequest. That is outside the new CreateScoresRequest = typing.List[CreateScoreRequest], and _assert_parsed expects a response type that no longer exists in the union.

  Verification: Trigger: any pytest run that collects tests/unit (CI included). Line 22 of tests/unit/test_scores_batch_response.py is still from langfuse.api.scores.types.create_score_response import CreateScoreResponse, so collection raises ModuleNotFoundError and the whole module errors. As a result, the sync and async batch-202 and batch-207 parsing tests in this file stop running.