Skip to content

fix(dashboard): open the row the Skills, Steering and Prompts filters name - #24771

Merged
bolichen97 merged 1 commit into
mainfrom
fix/skills-filter-selects-best-match-20262
Oct 27, 2026
Merged

bolichen97 merged 1 commit into
mainfrom
fix/skills-filter-selects-best-match-20262

Conversation

@bolichen97

@bolichen97 bolichen97 commented Oct 27, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: When you type in the Skills or Steering filter, the page opens the row you typed, not just the first row it shows.

On Customize › Skills, typing "widgets" left two rows, Artifacts and Widgets, and opened Artifacts on the right. Artifacts is listed because its description mentions widgets. The user had to stop and check which skill they were looking at. The Steering tab does the same.

Why it matters

The filter is how people find one skill or steering file. Opening a different one makes the filter look broken and slows every first-time user down; the nightly GUI user test hit it three times.

Not a goal

  • Changing what the filter matches: description matches stay listed.
  • Changing the list order.
  • Changing which row opens when the filter is empty.

What changed (motivation → approach → change)

The Skills, Steering and Prompts tabs open a row on their own so the right pane is never empty. That auto-pick always took the first visible row, and each filter also matches descriptions, so the first row can be one that only mentions the query.

Now the auto-pick takes the best match for the filter text:

Filter has Opens
a row whose name or key equals it that row
a row whose name or key contains it the first such row
only description matches the first row, as before

A row the user clicked stays open while the filter still shows it. The shared rule lives in website/src/utils/filterSelection.ts; SkillsTab.tsx, SteeringTab.tsx and PromptsTab.tsx call it.

Backwards compatibility

Compatible: the list, its order and the empty-filter default are unchanged; only which row opens by itself while filtering moves.
Removes nothing: no row, field, route or key is hidden or removed.

Tests

  • website/src/test/SkillsTabFilterSelection.test.tsx: exact and partial name matches open the typed skill; a clicked skill stays open; a clicked skill that the filter hides moves to the best match; description-only matches keep the first row. Three of the five fail on main.
  • website/src/test/SteeringTab.test.tsx: the same two cases for Steering; the name-match case fails on main.
  • website/src/test/PromptsTab.test.tsx: the same two cases for Prompts; the name-match case fails on main.

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

Manual verification

Reverted the two components to main and re-ran the new tests: the name-match cases fail, the rest pass. Ran the Skills, Steering, Prompts, Capabilities and list-detail test files (339 tests), tsc -p tsconfig.app.json and eslint on the changed files.

Screenshots / video

Customize › Skills with "widgets" typed. The API is stubbed with three skills; Artifacts' description mentions widgets.

Before (main): Artifacts opens.

before

After: Widgets opens; Artifacts stays listed.

after

Related Issues

Closes #20262

Sibling sweep

Swept: git grep -nE "set(Selected|Active)[A-Za-z]((all)?[Ff]iltered[A-Za-z][0]|(all)?[Ff]iltered[0]" -- website/src/pages
Sibling: website/src/pages/overview/SteeringTab.tsx:584 -- changed here
Sibling: website/src/pages/overview/PromptsTab.tsx:432 -- changed here
Sibling: website/src/pages/chat/crew-window/crewWindowAgentMode.tsx:100 -- not affected -- Enter picks the row only when exactly one row matches
Sibling: website/src/pages/chat/crew-window/crewWindowModelPicker.tsx:86 -- not affected -- Enter picks the row only when exactly one row matches
Sibling: website/src/pages/overview/AgentTemplatesTab.tsx:604 -- not affected -- it re-picks only when the selection leaves the unfiltered list, so typing a filter never moves it

Pattern harvest

Not generalizable: the defect is one auto-select rule shared by three list tabs, and all three now use one helper.

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

@bolichen97
bolichen97 requested review from a team and KiroCrewCaptain as code owners October 27, 2026 06:43
@bolichen97
bolichen97 requested a review from mbriones98 October 27, 2026 06:43
@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 27, 2026
@github-actions

github-actions Bot commented Oct 27, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of 7dddc0de302d231a730c0900e37f0845e6404161 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks: one shared helper fixes the auto-pick rule in all three tabs, and the screenshots match the stated behaviour.

Notes

  • The Prompts auto-pick now calls load(best) whenever the best match changes as the user types, so typing a name can fetch several prompt details in a row. It is cheap and easy to revert, but debouncing it or deriving the pick lazily would avoid the extra fetches.

[DESIGN-REVIEWED] 7dddc0d

Description read

[DESCRIPTION-READ] 971ea93e550fd33a5b6c5441f8f37e0f4a9b188e6cc6b5eb2b96323d59ebeab9

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 Oct 27, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of 7dddc0de302d231a730c0900e37f0845e6404161 — 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. All three tabs now use one helper, so the fix covers the cause, the auto-pick rule they shared, and every sibling is accounted for.

Notes

  • Item 5 — oversized: bestFilterMatch adds a third way to rank rows by a query, alongside the plain substring filter and website/src/utils/fuzzyMatch.ts, which already "scores highest" on exact matches (fuzzyMatch.test.ts:48). The helper has three consumers, so it is not speculative, but the overlap is real. Subtraction: none here. Swapping in fuzzyMatch would change the description-only fallback, which ## Not a goal keeps.

What this change ships

Inventory (6 items) — 5 justified

Intent: make typing a name in a Customize list filter open that row instead of the first row whose description mentions it. FIX (#20262).

  1. Skills filter opens the row whose name equals or contains the query — justified
  2. Steering filter does the same — justified
  3. Prompts filter does the same — rides along (the Goal names only Skills and Steering; this is the swept sibling)
  4. A row the effect opened on its own now moves when the filter changes; a row the user clicked stays while it is still visible — justified
  5. New shared helper bestFilterMatch in utils/filterSelection.ts — oversized (overlaps utils/fuzzyMatch.ts; 3 consumers)
  6. A newly created steering file counts as one the user clicked — undeclared

[FIRST-PRINCIPLES-REVIEWED] 7dddc0d

Description read

[DESCRIPTION-READ] 1b97f821eac2b103f4524264a0f42295b71fdf9f65c01d82ddd9a3f817e78aca

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 Oct 27, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ✅ PASS

UX-level review of 7dddc0de302d231a730c0900e37f0845e6404161 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

UX-Verdict: PASS

Nothing blocks: the PR adds no controls or copy, and it only changes which row opens by itself. The before and after screenshots show that row.

Notes

  • In the "before" shot, the blind reader asked why "Artifacts" opened when they had typed "widgets". That is the bug this PR fixes, and in the "after" shot Widgets opens.
  • The reader could not tell the Artifacts and Widgets detail panels apart: both show the same toggle and "0B · not delivered…" line. The fix moves the selection with no visible cue. A follow-up could make the detail heading stand out more so the change of row is easier to see.

[UX-REVIEWED] 7dddc0d

Description read

[DESCRIPTION-READ] 5128bdfaa07317a22adcd0d2e8864423c0b23c6352cb1cc9b44eca3aa1d8cddd

That sha256 names everything this verdict read: the PR description, plus the 5 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 Oct 27, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed 7dddc0de302d231a730c0900e37f0845e6404161 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 7dddc0d

Verdict parsed from the review's SHA-scoped output markers for commit 7dddc0de302d231a730c0900e37f0845e6404161.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 7dddc0de302d231a730c0900e37f0845e6404161 false-positive: <reason, 20 to 500 characters>

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 27, 2026
@github-actions

github-actions Bot commented Oct 27, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of 7dddc0de302d231a730c0900e37f0845e6404161 and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 7dddc0d

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 7dddc0de302d231a730c0900e37f0845e6404161 false-positive: <reason, 20 to 500 characters>

@bolichen97 bolichen97 changed the title fix(dashboard): open the row the Skills and Steering filters name fix(dashboard): open the row the Skills, Steering and Prompts filters name Oct 27, 2026
@bolichen97
bolichen97 force-pushed the fix/skills-filter-selects-best-match-20262 branch from 0dad77d to 86ed694 Compare October 27, 2026 06:56
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 27, 2026
@bolichen97
bolichen97 enabled auto-merge (squash) October 27, 2026 07:09
@bolichen97
bolichen97 force-pushed the fix/skills-filter-selects-best-match-20262 branch from 86ed694 to cba4560 Compare October 27, 2026 07:23
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main fixed the Integration red (/api/publish-providers NameError, #24767) since this head went up (86ed694 -> cba4560); CI re-running.

@bolichen97

Copy link
Copy Markdown
Contributor Author

Backend Tests (Windows) (2) fails on test/test_crew_log_fullscan_guard.py::test_a_warm_session_read_that_earns_a_savepoint_reads_only_the_gap with XPASS(strict). This PR touches no backend code; the strict xfail is stale on main and #24774 removes it. Waiting for #24774, then rebasing.

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 27, 2026
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased onto main with #24774 (stale strict xfail removed) (cba4560 -> 4cbdf10); CI re-running.

@bolichen97
bolichen97 force-pushed the fix/skills-filter-selects-best-match-20262 branch from cba4560 to 4cbdf10 Compare October 27, 2026 12:57
@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 27, 2026
… name

The auto-pick took the first visible row, and both filters also match
descriptions, so typing "widgets" opened Artifacts. It now opens the best
name or key match; a row the user clicked stays open while still listed.

Closes #20262
@bolichen97

Copy link
Copy Markdown
Contributor Author

Rebased: main moved in website/src/pages/overview (PortabilityTab) since this head went green (4cbdf10 -> 7dddc0d); the old head's Windows shard 8 sat queued for 2h with no runner. CI re-running.

@bolichen97
bolichen97 force-pushed the fix/skills-filter-selects-best-match-20262 branch from 4cbdf10 to 7dddc0d Compare October 27, 2026 15:21
@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: checking Automated validation is still running labels Oct 27, 2026

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

Tier 1 auto-approve: fix (7 files). Criteria: no conflict, no requested changes, security path denylist clean, design-doc gate clean, security checklist all-NO, no reviewer-lane findings, readiness signal fresh for this head SHA, declared category matches the diff. Category: dashboard Skills, Steering and Prompts tabs open the best filter match unless the user picked a row, plus tests.

@bolichen97
bolichen97 merged commit 7eeb297 into main Oct 27, 2026
108 checks passed
@bolichen97
bolichen97 deleted the fix/skills-filter-selects-best-match-20262 branch October 27, 2026 17:06
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Oct 27, 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.

ux(capabilities): After filtering for "widgets", the app had already selected and opened "Artifacts" instea…

2 participants