Skip to content

chore: pin pr-reviewer-plugin v0.36.0 — the QA panel check run - #49

Merged
mabry1985 merged 1 commit into
mainfrom
chore/pin-pr-reviewer-0360
Aug 24, 2026
Merged

chore: pin pr-reviewer-plugin v0.36.0 — the QA panel check run#49
mabry1985 merged 1 commit into
mainfrom
chore/pin-pr-reviewer-0360

Conversation

@mabry1985

Copy link
Copy Markdown
Member

Picks up pr-reviewer-plugin#97: the panel publishes a QA panel check run from the same decision that drives approve-on-green.

Why it matters for Vera specifically: an App's approval never satisfies a required approving review (GitHub counts reviewers with write access; an App is not one), so her verdict could not stop a merge no matter how it was posted — a WARN with open findings and a clean PASS were equally unenforceable. A check run is a first-class required status.

  • clear verdict, findings resolved → success
  • unresolved review threads → failure (the WARN gate)
  • FAIL standing against this head → failure
  • no verdict / stale head / incomplete pass / CI pending or red → in progress

Her live posture already owns promotion (shadow_mode: false, promotion_owner: true) and the check rides that same gate, so she starts publishing on this roll. It stays inert until a repo's ruleset lists QA panel as a required check.

Prerequisite: the protoreview App needs Checks: read & write. Without it the write logs a warning naming that permission and nothing else changes.

Manifest pin, Dockerfile bake ARG and the README member table move together — CI asserts all three agree.

🤖 Generated with Claude Code

v0.36.0 publishes a `QA panel` check run from the same decision that drives
approve-on-green (pr-reviewer-plugin#97). It is the first form of Vera's verdict
that GitHub will gate a merge on: an App's approval never satisfies a required
approving review (it counts reviewers with write access, and an App is not one),
so until now a WARN with open findings — or a FAIL — could not stop a merge.

Vera's live posture already owns promotion (`shadow_mode: false`,
`promotion_owner: true`), and the check rides that same gate, so she starts
publishing it for her repos on this roll. It stays inert until a repo's ruleset
lists `QA panel` as a required status check.

Needs the protoreview App granted `Checks: read & write`; without it the write
logs a warning naming that permission and nothing else changes.

Bundle manifest, Dockerfile bake ARG and the README member table move together —
CI asserts all three agree.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

@protoreview protoreview Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

QA panel review — PASS

code-review-structural · head f25943bf8efe · formal

PR #49 is a clean three-file version-pin bump (v0.35.0 → v0.36.0) for pr-reviewer-plugin: Dockerfile ARG, README table, and bundle manifest are all internally consistent. The verifier confirmed no defects. No prior requests to disposition. No panel disagreement. No coverage gaps — the change is small and fully walked.

No findings — the review came back clean.

findings JSON (machine-readable)
[]

@mabry1985
mabry1985 merged commit 4ab7766 into main Aug 24, 2026
1 check passed
@mabry1985
mabry1985 deleted the chore/pin-pr-reviewer-0360 branch August 24, 2026 02:12
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