[teamleaderleo]

Leo, on Swift suites that leave windows open for the suites after them: "But why do you guys not just clean up after yourselves, or at least do your best to?"

The first batch of suite fixes on top of #27003 (the window census and the .closesItsWindows check): the 20 sidebar suites in CmuxNextAppTests that the census named as leaking. Each now closes every window it opens in its own teardown, and carries .closesItsWindows, so a new leak fails that suite by name.

What leaked and why

• 16 suites ended their tests with window.orderOut(nil), which hides a window but leaves it open in the app's window list: SidebarBackgroundChangeCalm, SidebarCalmMotion, SidebarChannelsAfterHome, SidebarCloseWhileScrolled, SidebarFocusedVisible, SidebarInsertWhileDisplayAsleep, SidebarListTilePress, SidebarNewRowInPlace, SidebarPinnedRowWidth, SidebarRevealSelection, SidebarRowHoverFollow, SidebarRowHoverGap, SidebarRowHoverInset, SidebarSelectionRevealsTop, SidebarUniformRowSpacing. Now close() (with isReleasedWhenClosed = false where the test made the window).
• SidebarDoneFoldTests: one test never closed its window.
• SidebarRowsMoveOnRenderServerTests: its restore hid the window, and its "needs a screen" check threw before any teardown ran.
• SidebarMenuAnchorTests: the group editor's hide() keeps its panel for the next open (deliberate reuse), so the tests now close that panel too.
• SidebarOneMorphingCardTests: hover card panels stayed as child windows; teardown closes them.
• SidebarSearchOpensCenteredTests: the palette's hide() keeps its panel (deliberate reuse); teardown closes it.

No product code changes. The row-edge assertion now checks the documented 1.5 pt contract; its old minimum-width assertion contradicted that behavior.

Performance

At head bceb24c8e54cdddc0d857a838df4c72d42bf6027 versus base 39bb822b0f83b92593e93f542b71bfd61fbe1150, the same 20 suite filters each completed with a final summary and exit census on both sides.

metric	before	after	delta	harness or run link
Open windows at process exit, summed across 20 filters	84	0	-84	package-test-lane.sh suite; base step 8b4c36d7d222b3a7bb1008c8, head step 038fdffb
Closed but still alive windows, summed across the same filters	24	108	+84	same runs

Observed / Not run

The focused head run completed all 96 test definitions: 19 suites passed; SidebarOneMorphingCardTests completed with two failures. Both goingFromAWorkspaceCardToTheUpdateCardMorphsTheOneCard and goingFromTheUpdateCardToAWorkspaceCardMorphsBackAndParksTheHighlights fail the same size assertions with empty animated-size arrays on base. The shared classifier reports no new focused failure. SidebarRowHoverFollowTests now passes.

Independent source review passed at this head, and the fleet Release compile passed. The broad Swift job was killed after 1500 seconds with exit 143, ten unfinished tests and no final summary. It recorded 1,056 WindowServer-owned windows. That incomplete run is a validation blocker; the focused improvement does not establish full-package cleanup. Complete paired broad validation is pending.

The two No new SwiftUI findings are inherited: the offending ChatColors files, guard and allowlist match base. The sidebar-record warning entered through the CI merge tree. Neither classification waives the incomplete runtime run.

Risk: test-only.

Changelog

none

Agent session: agent session

[cursor]

Bugbot needs on-demand usage enabled

Bugbot uses usage-based billing for this team and requires on-demand usage to be enabled.

A team admin can enable on-demand usage in the Cursor dashboard.

[teamleaderleo]

Owned by an agent session · session

[github-actions]

cmux status _{(one comment, edited in place; each line is updated by the job that owns it)}

Review bceb24c: HQ feed card · gallery before/after runs · checks

App of this head (app.zip, kept 1 day)

gh workflow run reload-build.yml -R manaflow-ai/cmux --ref feat-cmux-next -f tag=pr27126 -f ref=window-leak-fix-A -f platform=macos -f backend_mode=local

red: cmux-next checks (god files, concurrency, crash safety, l10n); no finished base run on 39bb822 to compare; do not merge until green; see lane-rules

[teamleaderleo]

Bead cx-ap79p0 created for this PR by the cmux coordinator.

[teamleaderleo]

Review (codex gpt-6-astra high, next-review): SHIP at 720a050

No blocking findings.

[teamleaderleo]

Owned by an agent session · session

[teamleaderleo]

Red-by-red base comparison for 720a050:

• BackgroundAccessoryPolicyTests, BackgroundWork (ControlTests), CmdNRowSlot, NumberedShortcutPage, PaletteOpenLatency, SidebarCmdNumberTwice, SidebarFooterAlignment, SidebarHoverCardRetargetBench, SidebarTabMiddleClickReopen, TerminalRevealFrame, and WorkspaceGroupPinLive were compared with the same harness at base cb7adb2 and the merged-base A tree. NumberedShortcutPageTests (7 issues), SidebarCmdNumberTwiceTests (2), and SidebarFooterAlignmentTests (18) failed identically on base and A; every other named suite passed on both. These are base reds.
• The fleet comparison had no crash and the timing-sensitive PaletteOpenLatency suite passed at base and A. No PR-owned red remains.

The broad window run after this PR reported 0 windows open at exit (fleet step e0afb44a3cdb38f7111f3f9f9).

[blacksmith-sh]

Found 6 test failures on Blacksmith runners:

Failures

Test	View Logs
/tmp/tmp5imnlely/summary-fail.json (live):/summary-fail.json (live):	View Logs
/tmp/tmp939qzrsu/summary-missing.json (live):/summary-missing.json (live):	View Logs
[new_workspaces] debug.new_workspace timed 0 of 50 new workspace rows/ [new_workspaces] debug.new_workspace timed 0 of 50 new workspace rows	View Logs
[new_workspaces] the window draws no frames/[new_workspaces] the window draws no frames	View Logs
[tabs] debug.tab_switch timed 0 of 40 switches/ [tabs] debug.tab_switch timed 0 of 40 switches	View Logs
[tabs] the window draws no frames/[tabs] the window draws no frames	View Logs

^{Need help on this PR? Tag @codesmith-bot with what you need.}

[teamleaderleo]

Review (codex-review-box/gpt-6-astra high, lane-place): SHIP at bceb24c

SHIP

No actionable findings. Window cleanup and the 1.5 pt expectation match the current implementations. AppKit tests could not run in this Linux box without Swift.

[cmux-bot]

cmux-bot approves bceb24c: an independent review passed this exact head (review box review-27126-r2 receipt by teamleaderleo). A push needs a new review; a later BLOCK at this head dismisses this approval.

[github-actions]

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.4099263986101 (1816f75ed1)

[github-actions]

Shipped in:

• cmux NEXT nightly 1.0.0-nightly.4099184203401 (0b45ce8ec3)
• cmux NEXT nightly 1.0.0-nightly.4099502414501 (b02c564547)
• cmux NEXT nightly 1.0.0-nightly.4099558415201 (c0331557f5)