Skip to content

fix(a11y): keep focus on the Skills and Steering refresh while it runs - #32168

Closed
RohanK6 wants to merge 1 commit into
mainfrom
crew/conductor/r510-library-refresh-focus
Closed

RohanK6 wants to merge 1 commit into
mainfrom
crew/conductor/r510-library-refresh-focus

Conversation

@RohanK6

@RohanK6 RohanK6 commented Nov 7, 2026

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: pressing Refresh on the Skills or Steering tab from the keyboard keeps focus on the button while the refresh runs, instead of dropping it to the page.

Both icon Refresh buttons took a native disabled while their refetch ran. A disabled button leaves the tab order, so focus fell to the page for the whole refresh.

Why it matters

A keyboard or screen-reader user who refreshes loses their place and has to tab back from the top of the page every time.

Not a goal

What changed (motivation → approach → change)

Symptom: focus drops on Refresh. Cause: native disabled while busy. Change: website/src/pages/overview/SkillsTab.tsx and website/src/pages/overview/SteeringTab.tsx mark the busy button aria-disabled with the same dimmed style and guard the click, so a second press does nothing, the way the Ops Mission Control refresh does (#32049). The busy accessible name ("Refreshing skills…") is unchanged.

Backwards compatibility

Compatible: the button looks and is named as before; a second press while busy still does nothing.

Removes nothing.

Tests

  • website/src/test/SkillsTab.test.tsx and website/src/test/SteeringTab.test.tsx: the existing busy-refresh test now focuses the button, presses it, and checks it is aria-disabled and not disabled, keeps focus, ignores a second press, and keeps focus once the read lands. Both fail on main (the busy button was disabled). All 256 tests in the Skills, Steering and list-pane suites pass.

Proof: not provable -- frontend test (prove.py runs Python only)

Manual verification

N/A — unit coverage sufficient: the change is focus and ARIA state, covered by the tests above.

Why no screenshot: the busy button keeps the same dimmed look and label; only its focusability and ARIA state change.

Related Issues

Fixes #32162

Deferred: #32164 (the Retry and Load-more buttons on the other Capabilities and Memory tabs)

Sibling sweep

Swept: git grep -n "disabled={isFetching}" -- website/src/pages/overview
Sibling: website/src/pages/overview/SkillsTab.tsx -- fixed -- Refresh skills
Sibling: website/src/pages/overview/SteeringTab.tsx -- fixed -- Refresh steering files
Sibling: website/src/pages/overview/PromptsTab.tsx -- deferred -- Retry read, Deferred: #32164
Sibling: website/src/pages/overview/WorkflowLibraryTab.tsx -- deferred -- Retry, Deferred: #32164

Pattern harvest

Not generalizable: the remaining sites are listed in #32164.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)
  • 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

Contributed under the Apache-2.0 license of this project.

The Refresh buttons on the Skills and Steering tabs used a native
disabled while their refetch ran. A disabled button leaves the tab
order, so a keyboard press dropped focus to the page for the whole
refresh. Mark the busy button aria-disabled and ignore a second press
instead, the way the Ops Mission Control refresh already does.
@RohanK6
RohanK6 requested review from a team and KiroCrewCaptain as code owners November 7, 2026 07:12
@RohanK6
RohanK6 requested a review from tim-finnigan November 7, 2026 07:12
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Nov 7, 2026
@github-actions

github-actions Bot commented Nov 7, 2026

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of 1766582de99163467417d5c3968a4f1d2735fc48 — 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

Nothing blocks. Both Refresh buttons now use aria-disabled plus a click guard, the same inline pattern already used at many sites in the repo. The test changes pin the defect from #32162.

What this change ships

Inventory (4 items) — 4 justified

Intent: let a keyboard or screen-reader user press Refresh on the Skills or Steering tab without losing focus while it runs (FIX, #32162).

  1. Skills Refresh keeps keyboard focus and stays in the tab order while busy — justified
  2. Steering Refresh keeps keyboard focus and stays in the tab order while busy — justified
  3. A second press while busy still does nothing, now enforced by a click guard instead of the native disable — justified
  4. The busy button keeps the same dimmed look (aria-disabled:opacity-30) — justified

[FIRST-PRINCIPLES-REVIEWED] 1766582

Description read

[DESCRIPTION-READ] e090604167ee2beae8f8145db487e812de0ef4c15844cebf49cba2b2f861fff7

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 7, 2026

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of 1766582de99163467417d5c3968a4f1d2735fc48 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: the fix targets the real cause (native disabled while busy), matches the Mission Control refresh pattern from #32049, and adds no new visible surface.

Notes

[DESIGN-REVIEWED] 1766582

Description read

[DESCRIPTION-READ] dc46485a0e3848b3e6b288e6f5b4b19cfab79a98904569d0484c8f2935156310

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 7, 2026

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of 1766582de99163467417d5c3968a4f1d2735fc48 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

Nothing changes visually: the busy Refresh button stays at the same 30% opacity with the same label. The change is focus retention and ARIA state, and the tests cover both.

Notes

  • className="aria-disabled:opacity-30 aria-disabled:cursor-default" in SkillsTab.tsx and SteeringTab.tsx does not stop Btn's hover:border-border-strong hover:bg-bg-hover. While a refresh runs, the dimmed button still reacts to hover as if it can be clicked. Add aria-disabled:hover:border-border aria-disabled:hover:bg-transparent, as ChannelPage.tsx:365 does.
  • The cursor on the busy button changes from not-allowed to default, so it no longer matches other disabled Btn controls. Use aria-disabled:cursor-not-allowed to keep it as it was.

[UX-REVIEWED] 1766582

Description read

[DESCRIPTION-READ] aa50da4aae283f82d3346adbf0a194464ca4f2aab6416546fbe260754ab1b0ff

That sha256 names everything this verdict read: the PR description, plus the 3 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.

@RohanK6

RohanK6 commented Nov 7, 2026

Copy link
Copy Markdown
Contributor Author

Closing as a duplicate of #32163, which makes the same Skills and Steering refresh focus change and was opened first.

@RohanK6 RohanK6 closed this Nov 7, 2026
@github-actions

github-actions Bot commented Nov 7, 2026

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed 1766582de99163467417d5c3968a4f1d2735fc48 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 1766582

Verdict parsed from the review's SHA-scoped output markers for commit 1766582de99163467417d5c3968a4f1d2735fc48.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 1766582de99163467417d5c3968a4f1d2735fc48 false-positive: <reason, 20 to 500 characters>

@github-actions github-actions Bot removed the readiness: checking Automated validation is still running label Nov 7, 2026
@github-actions

github-actions Bot commented Nov 7, 2026

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of 1766582de99163467417d5c3968a4f1d2735fc48 and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 1766582

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 1766582de99163467417d5c3968a4f1d2735fc48 false-positive: <reason, 20 to 500 characters>

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.

Skills and Steering tabs: the Refresh button drops keyboard focus while it runs

1 participant