Skip to content

tests: exercise pull request review - #1

Closed
khoi wants to merge 2 commits into
mainfrom
test/supareview-smoke
Closed

tests: exercise pull request review#1
khoi wants to merge 2 commits into
mainfrom
test/supareview-smoke

Conversation

@khoi

@khoi khoi commented Aug 6, 2026

Copy link
Copy Markdown
Contributor

This temporary PR changes the documented draft and fork behavior so SupaReview has a clear issue to report. It tests the live pull request workflow and will not be merged.

Note

This PR is a smoke test only. Its README change is intentionally wrong.

@khoi khoi closed this Aug 6, 2026
@khoi khoi reopened this Aug 6, 2026
@khoi khoi closed this Aug 6, 2026
@khoi khoi reopened this Aug 6, 2026
@khoi khoi closed this Aug 6, 2026
@khoi khoi reopened this Aug 6, 2026
@github-actions

github-actions Bot commented Aug 6, 2026

Copy link
Copy Markdown

SupaReview

Findings

The only changed line documents behavior that the unchanged workflow does not provide: draft jobs are explicitly gated off, and fork runs lack the required secret and write-capable token. The patch should be reverted or accompanied by workflow changes that actually implement those claims.

Findings

  • [P2] Do not claim that draft pull requests are reviewed
    SupaReview reviews draft and fork pull requests. It gives both subagents the same diff context, runs them in parallel, deduplicates their findings, and posts one review comment.

    The changed sentence at README.md:24 contradicts the workflow shown immediately above it and the checked-in trigger: jobs.supareview.if is github.event.pull_request.draft == false. Draft pull requests therefore have their review job skipped; only a later ready_for_review event can cause a review. Restore the “skips draft” wording or change the workflow to implement the documented behavior.
  • [P1] Do not claim that fork pull requests are reviewed
    SupaReview reviews draft and fork pull requests. It gives both subagents the same diff context, runs them in parallel, deduplicates their findings, and posts one review comment.

    The changed sentence at README.md:24 also claims fork support without any corresponding workflow or authentication change. For an ordinary fork-triggered pull_request, repository secrets such as CODEX_BALANCER_API_KEY are not available and the token is read-only by default, while this workflow passes that secret to the review action and needs write access to publish a review. The existing workflow consequently cannot reliably run and post reviews for fork PRs; restore the “skips fork” wording or add a deliberately fork-safe implementation.

Review type: Full review

Reviewers: thermo-nuclear-review-subagent, thermo-nuclear-code-quality-review-subagent

@khoi khoi closed this Aug 7, 2026
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.

1 participant