Skip to content

fix: backport 4 user-facing fixes to 0.12.1 (batch late-4) - #33742

Merged
buluoray merged 4 commits into
release/0.12.0from
crew/slack/0.12.1-late-4
Nov 9, 2026
Merged

buluoray merged 4 commits into
release/0.12.0from
crew/slack/0.12.1-late-4

Conversation

@buluoray

@buluoray buluoray commented Nov 9, 2026

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: Ship 4 user-facing fixes that merged to main after the 0.12 cut to 0.12.1 users.

Each fix is a bug that 0.12 users can hit today. Three cherry-pick onto release/0.12.0 with no conflict; one needed a small hand-port, listed below.

Why it matters

0.12 users keep hitting these bugs until a patch release carries them. A KAS tool result that only mentions a refusal is read as a gateway refusal. A pptx-maker engine whose uv.lock was rewritten never repairs (#31183). Keyboard focus drops off the md-notebook Sync and Save buttons while they run. The subagent bar covers the context-usage popover.

Not a goal

  • No new behaviour on release: each commit is a git cherry-pick -x of the main merge commit; the one hand-port keeps release's own code and applies only that fix's change.
  • No prerequisite feature picked to make a conflict go away.
  • No CI, test-only, docs-only or dev-tooling commits.

What changed (motivation → approach → change)

The 4 fixes are applied in the order given, one cherry-pick -x of each main merge commit, on release/0.12.0 at d684523f65.

# PR Main merge commit What users hit
1 #33640 fb1cdd61cd a tool result that only mentioned a refusal was read as a gateway refusal
2 #33679 1d2cc02c85 a pptx-maker engine whose uv.lock was rewritten was never repaired (fixes #31183)
3 #33451 f2e2089fac keyboard focus left the md-notebook Sync and Save buttons while they ran
4 #31305 acce269a1b the subagent bar covered the context-usage popover

Hand-resolved hunks (#31305). Main's version of the context-usage control carries two props release does not have, contextUnknown (from #32831) and cacheSlot (from #31501). Each conflict keeps release's side and applies only #31305's own change:

File Hunk Resolution
website/src/components/ChatInput.tsx <ContextUsageControl …/> call drop ctxWrapRef; keep release's props (no contextUnknown, cacheSlot)
website/src/components/chat-input/ContextShelf.tsx ContextUsageControl signature drop ctxWrapRef; keep release's props
website/src/components/chat-input/ContextShelf.tsx trigger button take #31305's {...press} and onClick guard; keep release's contextTip(effectivePct)
website/src/test/ChatInputCoverage.test.tsx added tests keep #31305's portal test; leave out #32831's two contextUnknown tests

The popover component and keepFocusOffTrigger that #31305 relies on are already on release.

Prerequisites not picked:

Dropped: none.

#31305's commit is authored Ray Xu <buluoray@gmail.com>, the identity Ray's earlier release picks use (#30903, #30946, #30644), so the repo's leak-guard hook accepts it.

Backwards compatibility

Compatible: three commits are the main commit unchanged, already reviewed and merged there; the fourth applies only #31305's change to release's own component. Each main PR states its own compatibility.

Removes nothing: no field, API, key or config is removed. #31305 replaces the popover's hand-rolled outside-click listener with the shared Popover's own dismissal.

Tests

Run on this branch at bdc5393914.

Backend, in a clean Python 3.12 venv (uv venv -p 3.12, uv pip install -e . -r pyproject.toml --group dev), with env -u KIROCREW_NATIVE_SKILL_PROJECTION PYTHONNOUSERSITE=1 and one pytest run per file:

  • test/test_pptx_maker_engine.py: 50 passed
  • test/test_pptx_maker_engine_source.py: 76 passed
  • test/test_recovery_l1_chat_runner.py: 60 passed

Frontend, one vitest run per file:

  • src/test/ChatInputCoverage.test.tsx: 78 passed
  • src/test/agentSwitchNotice.pairedSurfaces.test.tsx: 5 passed
  • src/test/mdNotebookSettingsBusyFocus.test.tsx: 1 passed
  • src/test/pressActivationDropdowns.test.tsx: 21 passed
  • tsc -p tsconfig.app.json --noEmit: clean

Proof: not provable -- other: a cherry-pick batch of already-merged main fixes, each proven in its own main PR

Manual verification

N/A: each fix was verified in its own main PR; this batch carries them unchanged except for #31305's prop-only hand-port above.

Screenshots / video

From #31305's main PR; the portal change here is the same.

#31305 a

#31305 b

More from #31305

#31305 c

#31305 d

#31305 e

#31305 f

#33451 keeps focus on a busy button and changes no pixel.

Related Issues

0.12.1 backport batch late-4: #33640, #33679, #33451, #31305. Fixes #31183 on the release line through #33679.

Sibling sweep

Siblings: none -- a backport batch; each pick's own PR carried its sweep

Pattern harvest

Not generalizable: a release backport batch, not a new defect fix.

Checklist

  • Each commit is a cherry-pick -x of a main commit with a Conventional Commits title
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

Contribution License Agreement

Contributions are accepted under the project's Apache-2.0 license; see the Licensing section of CONTRIBUTING.md.

RohanK6 and others added 4 commits November 9, 2026 06:01
The recovery ladder's first rung reads every tool result for the
gateway's capacity refusal and its recoverable markers, and a turn
whose last result matches is retried with a "lack of gateway
capacity" notice. A short result from a call that succeeded, such as
a search that prints "BackendGone" or "-32001", matched too, so a
normal turn was retried up to three times. The stub and the gateway
send a refusal as a JSON-RPC error, which the harness reports as a
failed call, so skip results that completed or are still streaming.

(cherry picked from commit fb1cdd6)
An install that hit #29225 keeps a marker that matches the pin and a
venv, so provisioning skipped both the fetch and the venv build and the
banner never offered it. `uv run --frozen` then installs the rewritten
lock exactly and the sdpm server still fails to start.

The engine now also counts as installed only while its mcp-local/uv.lock
hashes to the lock inside the pinned archive (ENGINE_LOCK_SHA256). A
drifted tree reads as not installed, so the banner offers provisioning
again and that run re-fetches the tree and rebuilds the venv with
`uv sync --frozen`.

Closes #31183

(cherry picked from commit 1d2cc02)
…33451)

The MD Notebook Sync button and the Settings access-token Save button set
native disabled while their own request runs. A disabled button leaves the
tab order, so a keyboard user who presses one loses focus to the page body
until the request settles.

Use aria-disabled (plus aria-busy) and a busy guard in the click handler
instead of native disabled, matching the pattern the clone/attach button
already uses. Focus stays on the pressed button and a second press is
ignored while the first request is in flight. The muted look and cursor
stay in the existing inline styles; the Save button keeps its native
empty-field disable.

The two confirm-row buttons (remove-vault confirm, clear-token) unmount
when their request finishes, so they are left for a separate change.

Co-authored-by: kirocrew-agent <agent@kirocrew.local>
(cherry picked from commit f2e2089)
@buluoray
buluoray requested review from a team and KiroCrewCaptain as code owners November 9, 2026 06:11
@buluoray
buluoray requested review from Zedmor and removed request for a team November 9, 2026 06:11
@github-actions

github-actions Bot commented Nov 9, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of bdc53939141bbb11725226eb79cb95249f71f32b and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] bdc5393

False positive or not applicable? A repository writer can comment:
/ai-review override gpt bdc53939141bbb11725226eb79cb95249f71f32b false-positive: <reason, 20 to 500 characters>

@github-actions github-actions Bot added the readiness: action required A blocking check or review needs attention label Nov 9, 2026
@github-actions

github-actions Bot commented Nov 9, 2026

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of bdc53939141bbb11725226eb79cb95249f71f32b — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: PASS

Four already-merged fixes, cherry-picked to the release branch. Each one is at mechanism level and has a test or a linked issue behind it. No defaults or product shape change.

Notes

  • Item 5 — undeclared: the outside-dismiss now fires when the press is released, not on mousedown, and a keyboard-opened card now takes focus. Both come with the shared Popover. The description only says "replaces the popover's hand-rolled outside-click listener with the shared Popover's own dismissal".

What this change ships

Inventory (6 items) — 5 justified

Intent: give 0.12.1 users four bug fixes that already merged to main. This is a FIX.

  1. A tool result whose status is completed, in progress or pending is no longer read as a gateway refusal — justified
  2. A pptx-maker engine whose uv.lock was rewritten now reads as not installed and is fetched again (pptx-maker: provisioning does not repair an engine whose uv.lock was already rewritten #31183) — justified
  3. The md-notebook Sync and Save buttons keep keyboard focus while they run, and ignore a second press — justified
  4. The context-usage popover is portalled so the subagent bar no longer covers it — justified
  5. The popover now dismisses when a press outside is released, and moves focus into the card when opened from the keyboard — undeclared
  6. The ctxWrapRef prop and the hand-rolled mousedown listener are removed — justified

[FIRST-PRINCIPLES-REVIEWED] bdc5393

Description read

[DESCRIPTION-READ] 92034a490656dcd0062feb77f5a87f0c6af56388558688fefddd9bbcf36c14ef

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

github-actions Bot commented Nov 9, 2026

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of bdc53939141bbb11725226eb79cb95249f71f32b — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: each of the four backports is scoped to the fix it ports. The only thing removed is the internal ctxWrapRef, and no reader of it is left in the tree.

Notes

[DESIGN-REVIEWED] bdc5393

Description read

[DESCRIPTION-READ] 0348abf61faf3f8c42e14307b98ec1acfb934778c2a91fc1e23c25d0de0ab0e9

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Nov 9, 2026

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of bdc53939141bbb11725226eb79cb95249f71f32b — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

Nothing blocks. The popover now draws above the subagent bar, and no new strings are added. Shots 1/4 show the fix and shots 2/6 show the old bug. The md-notebook busy-focus fix changes nothing on screen.

Notes

  • In the PR description, label the screenshots "after" (shots 1/4) and "before" (shots 2/6). The blind reader saw "two stacking orders" and could not tell which one this PR ships.

[UX-REVIEWED] bdc5393

Description read

[DESCRIPTION-READ] 6b614c59dda3c461313abaa3858813ed964c24ad80e393c82f0afd98d6c435e4

That sha256 names everything this verdict read: the PR description, plus the 9 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

github-actions Bot commented Nov 9, 2026

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed bdc53939141bbb11725226eb79cb95249f71f32b — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] bdc5393

Verdict parsed from the review's SHA-scoped output markers for commit bdc53939141bbb11725226eb79cb95249f71f32b.

False positive or not applicable? A repository writer can comment:
/ai-review override fable bdc53939141bbb11725226eb79cb95249f71f32b false-positive: <reason, 20 to 500 characters>

@buluoray
buluoray merged commit 34898f9 into release/0.12.0 Nov 9, 2026
20 of 30 checks passed
@buluoray
buluoray deleted the crew/slack/0.12.1-late-4 branch November 9, 2026 07:30
@github-actions github-actions Bot removed the readiness: action required A blocking check or review needs attention label Nov 9, 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.

4 participants