ref(evals): Centralize fixture provenance#329
Merged
Conversation
Move GitHub fixture path handling behind a shared eval fixture helper so scaffold, runner, and verification use the same provenance rules. Store scaffolded fixtures under repository-qualified paths while keeping eval repo paths and report output repo-relative. Record skipped scaffold files in notes so fixture gaps are visible. Co-Authored-By: GPT-5 Codex <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Keep scaffolded eval fixture provenance explicit without leaking it into eval output. GitHub-sourced fixtures now encode owner/repo under the checked-in fixture path, while runners and verifier map them back to repo-relative paths and prompt repository metadata.
Fixture Interface
Add a shared
src/evals/fixtures.tshelper for building fixture storage paths, deriving repo-relative paths, and detecting a single source repository. This keeps scaffold, runner, and verifier on one layout contract.Scaffold Visibility
Record skipped files and source repository in scenario notes, and print skipped files from
pnpm evals:scaffoldso base/head-only gaps are visible when a fixture is generated.