Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
I noticed that the docs are not up to date with the v0.1.8 API (see release notes).
In this PR, I'm updating the docs to reflect the changes (use
@vf.stop, follow updatedenv_responsesignature, ...)This also fixes #608.
@willccbb @mikasenghaas
Note that I did not update api_reference.md because it's unclear to me what should be exposed there. I'd prefer if a maintainer could handle this separately.
Type of Change
Testing
uv run pytestlocally.Checklist
Additional Notes
Note
Updates documentation to align with v0.1.8 protocol and state changes.
MultiTurnEnvexamples toasync env_response(...) -> Messages(no tuple) and use@vf.stop-decorated stop conditions instead of overridingis_completedstate["trajectory"],completion,is_completed,stop_condition, timing) and referenceTrajectoryStepstructureenv.rollout(input=...)usage,Rubric.score_rollout(state, ...), and scoring outputs now understate["reward"]/state["metrics"]; clarifysave_everybehaviorWritten by Cursor Bugbot for commit fd3d8d6. This will update automatically on new commits. Configure here.