You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
fix(source-control): pull-request merge must confirm the PR entered the merge queue #6630
Observed 2026-10-07: gh pr merge <n> --squash on a merge-queue branch prints "The merge strategy for main is set by the merge queue". Sometimes the PR still does not enter the queue: #6481 sat CLEAN, with auto-merge armed and isInMergeQueue false, until it was queued directly with the GraphQL enqueuePullRequest mutation. #6426 needed the same check.
Understood: gh pr merge --squash on a merge-queue branch can report success without the PR actually entering the queue (seen on #6481 and #6426), and the merge phase doc (plugins/source-control/skills/pull-request/reference/merge.md) doesn't verify isInMergeQueue or fall back to the enqueuePullRequest GraphQL mutation when it's false. The request is scoped clearly: after merge, check isInMergeQueue, enqueue directly when CLEAN-but-not-queued, and surface GitHub's enqueue error verbatim (e.g. an unresolved-conversation blocker) rather than failing silently. Acceptance criteria are already stated in the issue. No duplicates found in open issues. Nothing appears missing to act on this.
The merge phase gains §4.2.2 "Confirm queue entry": after gh pr merge on a merge-queue base it reads isInMergeQueue, enqueues a CLEAN PR left out once with enqueuePullRequest at the head pinned in 4.1, and reports GitHub's refusal text verbatim, so a run ends merged, queued, or held with a named blocker. 4.2 and 4.2.1 point at it, and an armed auto-merge is reported.
Verified with bash -n, shellcheck, a stubbed-gh run across 7 states, scripts/affected-tests.sh --base origin/main --run (plus the 2 selected Python suites, 48 passed), markdownlint and the changelog fragment checks.
Left for a human: no live run against a real merge queue (GraphQL is refused from this session). Marking the PR ready waits for Kyle.
Notes: Not run against a real merge queue (GraphQL 403 here). Defaults: re-run once if already queued; armed auto-merge reported, not disabled
Next step: Review the draft PR; decisions for Kyle are listed in its body. Mark it ready once Kyle agrees (that starts the CI test and review lanes), then drive CI green.
Where a branch is listed it is pushed, so it can be taken over as is. A draft PR shows a red ci-status by design until it is marked ready.
Activity
github-actions commentedon Oct 8, 2026
Understood:
gh pr merge --squashon a merge-queue branch can report success without the PR actually entering the queue (seen on #6481 and #6426), and the merge phase doc (plugins/source-control/skills/pull-request/reference/merge.md) doesn't verifyisInMergeQueueor fall back to theenqueuePullRequestGraphQL mutation when it's false. The request is scoped clearly: after merge, checkisInMergeQueue, enqueue directly when CLEAN-but-not-queued, and surface GitHub's enqueue error verbatim (e.g. an unresolved-conversation blocker) rather than failing silently. Acceptance criteria are already stated in the issue. No duplicates found in open issues. Nothing appears missing to act on this.Automated triage. A maintainer reviews it.
kyle-sexton commentedon Oct 11, 2026
Draft PR: #7075 (branch
fix/6630-merge-confirm-queue-entry).The merge phase gains §4.2.2 "Confirm queue entry": after
gh pr mergeon a merge-queue base it readsisInMergeQueue, enqueues aCLEANPR left out once withenqueuePullRequestat the head pinned in 4.1, and reports GitHub's refusal text verbatim, so a run ends merged, queued, or held with a named blocker. 4.2 and 4.2.1 point at it, and an armed auto-merge is reported.Verified with
bash -n, shellcheck, a stubbed-ghrun across 7 states,scripts/affected-tests.sh --base origin/main --run(plus the 2 selected Python suites, 48 passed), markdownlint and the changelog fragment checks.Left for a human: no live run against a real merge queue (GraphQL is refused from this session). Marking the PR ready waits for Kyle.
Generated by Claude Code
fix(source-control): pin merges to the head captured in 4.1 and warn …
kyle-sexton commentedon Oct 11, 2026
Handoff note for whoever picks this up next. The overnight Claude run has stopped, and Kyle is moving issue work to Cursor.
fix/6630-merge-confirm-queue-entryWhere a branch is listed it is pushed, so it can be taken over as is. A draft PR shows a red
ci-statusby design until it is marked ready.Generated by Claude Code