[ggbdpq]

ELI5

The sidebar's Workspace options button shows a little number badge while any workspace filter is active. If you keep filters on purpose (many people do), that badge never goes away and turns into permanent visual noise. This PR adds a "Hide filter badge" switch in the same menu, so you can keep your filters and lose the badge.

What Changed

• New persisted preference hideWorkspaceFilterBadge (default false — current behavior is unchanged), wired through the same six persistence seams as the neighboring workspace-filter preferences: shared PersistedUIState type, UI slice contract, preference actions, hydration, the persisted-UI write baseline (type + field set), and the persisted-UI writer mirror.
• A "Hide filter badge" FilterToggleRow at the end of the Filters section of the Workspace options menu.
• The badge render condition becomes hasAnyFilter && !hideWorkspaceFilterBadge. The button's accessible name and tooltip still report the active-filter count while the badge is hidden — screen-reader users keep the state; only the visual noise goes.

Why

The issue offers two directions: a setting to hide the badge, or a subtler dot. We implemented the toggle because it fully removes the noise for people who intentionally run filters long-term, and placing it inside the Workspace options menu — right where filters are turned on — makes it discoverable exactly where the annoyance comes from, without touching the Settings window. The dot variant can still be layered on later if the maintainers prefer it. No changes to the filter-count logic itself: the toggle is a display preference, not a filter, so it never counts toward the badge number.

Linked Issue

Fixes #25127

Visual Proof

N/A for attached media in this environment (headless run). Behavior is pinned by the new component tests: the badge renders with the default preference, disappears while "Hide filter badge" is on, and the accessible name keeps the active-filter count in both states.

Testing

• I manually tested these changes locally
• Automated tests added/updated, or explained why not below

New SidebarWorkspaceOptionsMenu.test.tsx (3 tests, red before the fix — the hide case failed with expected '2' to be null — green after). Also updated persisted-ui-write-baseline.test.ts fixture (typecheck-enforced).

Commands actually run locally (macOS): pnpm tc (exit 0), target + baseline + hydration suites via pnpm test (13 passed in the re-run), adjacent store/sidebar suites (123 passed across 7 files), npx oxlint on all touched files (0 warnings, 0 errors), verify:localization-catalogs / verify:localization-coverage (exit 0). Full pnpm test / pnpm build not run locally — CI covers them.

One incidental change: en-runtime-required.json had sendOutcomeLost out of alphabetical order (upstream commit landed without running the sync script); sync:localization-runtime-catalog moved it back — content identical, and the catalog checks fail without it.

AI Disclosure

Coding by GLM-5.3-Flash (ZCode agent), orchestrated and verified by a main agent with the same model; a human reviewed and committed the result.

Agent skill upstream boundary

• Not applicable, or this change follows docs/reference/agent-skill-sharing-upstream-boundary.md and copies or mechanically translates no upstream skill-installer source, tests, fixtures, registry entries, path tables, comments, or documentation.

Notes

Pure renderer UI change: no platform branches, no SSH/remote surface, no wire/compat impact, no new dependencies. Cross-client persistence is additive — the new field is optional in PersistedUIState, and older clients ignore unknown keys per the shared-state convention.

Checklist

• This PR is small and focused
• I explained what changed and why (ELI5, the user-facing before/after, the mechanism, and why over the alternatives)
• Before/after screenshots or videos attached for UI changes, or N/A with reason
• Self-reviewed for correctness, security, and performance
• Cross-platform, SSH/remote, and path/shortcut impact considered (or N/A)
• pnpm lint, pnpm typecheck, pnpm test, and pnpm build pass (or CI will cover; local preferred)

[pullfrog]

Caution

hideWorkspaceFilterBadge is added to PersistedUIState but not to the strict ui.set schema (UiUpdateFields), so pnpm tc fails and paired web/mobile/relay clients will reject the writer's payload. See the inline comment for the one-line fix.

Reviewed changes

• New persisted preference hideWorkspaceFilterBadge (default false) — optional field on PersistedUIState, wired through the UI slice contract, preference actions, hydration (=== true), the persisted-UI write baseline (type + field set), and the writer mirror.
• Badge suppression in SidebarWorkspaceOptionsMenu.tsx — render condition becomes hasAnyFilter && !hideWorkspaceFilterBadge; the accessible name and tooltip keep reporting the active-filter count.
• New FilterToggleRow in SidebarWorkspaceFilterSection.tsx — "Hide filter badge" with an EyeOff icon, appended to the Filters section.
• Tests + i18n — new SidebarWorkspaceOptionsMenu.test.tsx (3 tests, verified passing under config/vitest.config.ts); hideFilterBadge added to en.json; incidental alphabetical re-sort of sendOutcomeLost in en-runtime-required.json (content identical).

The persistence seams that are present are correctly wired; I verified the localization gates (verify:localization-catalogs, runtime catalog) pass and that the new test genuinely fails with the bug present. The one blocker is the missing companion schema entry below.

ℹ️ Nitpicks

• getDefaultUIState() and getStartupErrorFallbackUI() omit the new field. Harmless today (hydration reads === true, default false), but the neighboring hideDetachedHeadWorkspaces is listed there, so adding it keeps new profiles and the startup-error fallback consistent.

  ^{｜ Fix all ➔ ｜ Fix 👍s ➔ ｜ View workflow run ｜ Using deepseek-v4.1-flash (free via Pullfrog for OSS) ｜ 𝕏}

[coderabbitai]

Important

Review skipped

We couldn't safely recover the incremental review. No full review was started, and the last reviewed checkpoint was preserved. Retry later, or explicitly request a full review by commenting @coderabbitai full review.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• Trigger review

No actionable comments were generated in the recent review.

Walkthrough

Adds the hideWorkspaceFilterBadge preference to UI state and persisted UI data. A sidebar toggle controls the preference. When enabled, the active-filter count badge is hidden from the Workspace options button, while the count remains in its accessible name. Tests cover badge visibility and accessible naming.

Severity of issue fixed: Low

Merge Risk: Minimal · previous review 0dd7e

The preference hides only the visual badge while retaining the accessible filter count and is described as persisting across reloads. No concrete merge-blocking risk is evident; the change appears ready for normal checks.

Security architecture risk: Low · up to 0dd7e

The setting changes badge visibility without changing workspace filtering or removing the accessible filter count. No introduced security issue was established. Compatibility with older connected hosts remains unconfirmed and could prevent a bundled preference update from saving.

Retained concerns
No architecture-level concerns identified.

Security review details

Security Blast Radius

• inferred — The field is not classified as pairing-local, so the existing web UI-update path can carry it to the active host’s persisted UI state. The demonstrated new effect is shared badge presentation, not additional repository, credential, or execution authority.

Trust Boundaries and Controls

• observed — The new preference uses existing UI IPC and RPC update routes. At the reviewed head, the RPC schema constrains the field to a boolean and retains strict unknown-key handling; the changed menu code adds no new privileged invocation.

Resilience and Maintainability Implications

• observed — Existing writer failure handling settles in-flight markers without acknowledging rejected changes. Terminal rejection does not trigger an unconditional retry loop; applicable later edits can schedule another field diff. This contains transport failure but does not establish older-host compatibility.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 50.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 6 functions across 11 files. (1 skipped: …	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely describes the main change: adding a Hide filter badge toggle for workspace options.
Description check	✅ Passed	The description covers the user impact, implementation, rationale, linked issue, testing, and checklist. The required visual proof is not attached; the stated N/A reason does not apply because this ch…
Linked Issues check	✅ Passed	Issue #25127 requests a way to hide the active-filter badge without disabling filters. The PR adds the persisted hideWorkspaceFilterBadge preference, a Workspace options toggle, and a render conditi…
Out of Scope Changes check	✅ Passed	The UI, persistence, RPC schema, localization label, baseline update, and tests all support issue #25127. The reviewed change summary identifies no unrelated changes.

Full details: Docstring Coverage

Explanation

Docstring coverage is 50.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 6 functions across 11 files. (1 skipped: 1 unsupported.)

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

ℹ️ Recent review info

⚙️ Run configuration

• Configuration used: Repository UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 944514fd-5203-4691-a44c-ac38ec2dbd43

📥 Commits

Reviewing files that changed from the base of the PR and between ae6a886ba813cfeb4e9658fd77c5f1a57e582565 and 0dd7e6473ce874c5fa23b7cfbf33fcbc87d1a4eb.

📒 Files selected for processing (3)

• src/renderer/src/components/sidebar/SidebarWorkspaceOptionsMenu.tsx
• src/renderer/src/i18n/locales/en.json
• src/renderer/src/store/slices/ui/ui-slice-preference-actions.ts

🚧 Files skipped from review as they are similar to previous changes (1)

• src/renderer/src/i18n/locales/en.json

Included review availability: This review used your included allowance. Your plan provides up to 10 included reviews per hour; 8 remain after this review.

[ggbdpq]

Verified against the code — this was correct, thank you. Fixed in ae6a886ba8: hideWorkspaceFilterBadge: z.boolean().optional() now sits in ClientUiWorkspaceFilterFields, next to its neighbors. Reproduced red before the fix (tsc -b config/tsconfig.node.json --force → TS2322 ... { missingFromSchema: "hideWorkspaceFilterBadge" } at the parity assertion), green after, and swept the remaining consumers of the field for a second list to register in — none found.

One incidental finding worth flagging for reviewers: local pnpm tc stayed green the whole time — the incremental tsbuildinfo did not propagate the src/shared/persisted-ui-state-types.ts change to a re-check of ui-state-schema-parity-checks.ts, so only tsc -b --force surfaces the error. Not touching the build config in this PR, just noting it since the parity guard is the thing that makes this failure mode visible at all.

One semantic question for the maintainers: as written, the preference rides ui.set to paired devices (it is not in pairing-local-ui-fields.ts), so hiding the badge is a per-profile choice that syncs across desktop/web/mobile — same as the neighboring filter toggles. If the intent is instead that each device remembers its own badge visibility, the field should move into the pairing-local list. Happy to adjust either way.

[pullfrog]

✅ No new issues found.

The prior blocker is resolved: hideWorkspaceFilterBadge now spreads into the strict ui.set schema, pnpm tc:node passes, and the new component test stays green.

Reviewed changes (since the prior pullfrog review at 5493e59)

• Registered hideWorkspaceFilterBadge in the ui.set wire schema — added hideWorkspaceFilterBadge: z.boolean().optional() to ClientUiWorkspaceFilterFields, which spreads into the strict UiUpdateFields (client-ui-params.ts:163), satisfying the parity assertion and letting paired clients persist the preference.

  ^{｜ View workflow run ｜ Using deepseek-v4.1-flash (free via Pullfrog for OSS) ｜ 𝕏}

[ggbdpq]

Rebased onto current main (990b0ed): main extracted the inline badge into a shared <OptionsFilterCountBadge> component (#25376), so the hide toggle now gates that component instead of the inline span — same behavior, new structure (20 passed across the sidebar + persistence suites). One incidental drop: our earlier sendOutcomeLost alphabetical-order fix in en-runtime-required.json landed upstream with the same content, so that hunk is gone from the diff. Same CI heads-up as on #24689: pnpm tc has one pre-existing failure on clean upstream/main (structured-agent-session-stop-note-opened-turn.test.ts TS2353), unrelated here.

[ggbdpq]

Status update and gentle ping — no open review threads remain (the pullfrog wire-schema blocker was resolved in ae6a886), but the branch had drifted three weeks behind main.

• Rebased onto a886dcd today. Main's sidebar refactor (refactor(sidebar): the Workspace options menu renders from one shared option list #28931) removed SidebarWorkspaceFilterSection.tsx, so the toggle now lives in SidebarFilter.tsx next to the other hide-toggles, with the i18n key moved to the SidebarFilter locale block; the badge gating in SidebarWorkspaceOptionsMenu survived unchanged.
• Picked up one new upstream constraint: the WSL-move exhaustive UI-state check requires every new PersistedUIState key to carry a disposition, so hideWorkspaceFilterBadge is registered as a desktop-global field.
• Verification: SidebarWorkspaceOptionsMenu.test.tsx 3/3, FilterToggleRow.test.tsx + persisted-ui-write-baseline 14/14, wsl-move-dispositions.test.ts 5/5, typecheck:node and typecheck:web clean.

@nwparker gentle ping — happy to adjust anything.

[ggbdpq]

Rebased onto current main (1263370049) and force-pushed. One conflict area, resolved with both sides kept:

• Upstream's i18n pass (fix(i18n): translate remaining renderer leftovers (submodules, agent summaries, filter counts, automations) #35914) pluralized the options-menu filter tooltip/aria-label ({{count}} keys, dropped activeFilterLabel); the hide-toggle's store subscription now sits on top of that shape, and the test follows the pluralized label.

Verification: 4 targeted/adjacent suites pass (22 tests), typecheck:node and typecheck:web clean, code-quality gate 0 findings, oxfmt clean, max-lines ratchet clean.

@nwparker gentle ping — happy to adjust anything.

[greptile-apps]

Confidence Score: 5/5

[Medium impact] The PR appears safe to merge, though the existing request for an end-to-end switch-to-badge interaction test remains open.

Findings

1.  Switch interaction remains untested ▶

Summary

The PR adds a persisted preference that hides the visual workspace-filter badge while retaining the active-filter count in the button’s accessible name. Since the previous review, it moves the switch into the Workspace options menu and adds switch-interaction tests.

_{Reviews (2) · Last reviewed commit: "fix(sidebar): move the Hide filter badge..." · Reviewed by Greptile}

[ggbdpq]

Both points addressed in 57f22850d1:

• P1 (wrong menu) — confirmed and fixed. The switch was indeed rendered by SidebarFilter (Kanban drawer header), while the badge lives on the sidebar's Workspace options button. The toggle now renders as a HideFilterBadgeOptionItem at the end of the Filters section of the workspace options menu, reusing the existing menu row shape; SidebarFilter.tsx drops the toggle and its store subscriptions, and the i18n key moves with it (SidebarWorkspaceFilterSection.hideFilterBadge, value unchanged). The badge condition itself is untouched.
• P2 (interaction test) — added: the switch is asserted to render at the end of the Filters section, a click forwards to setHideWorkspaceFilterBadge(true), and aria-checked reflects the preference. Both new tests fail against the pre-move code. Focused suites 34/34, typecheck:node/typecheck:web, quality gate, oxfmt and the max-lines ratchet all clean.

[ggbdpq]

Rebased onto current main and force-pushed. Three notes: upstream's cleanup removed persisted-ui-write-baseline.test.ts (followed — this toggle's behavior coverage lives in the menu tests); the options-menu test moved with upstream's sidebar layout into worktree-list/controls/ and keeps the Hide filter badge toggle interaction regression block; and the deleted baseline test's fixture-seeding commit was dropped as it no longer had a home.

Verified after the rebase: options-menu + slice/sync suites 67/67, typecheck:node, typecheck:web, format check, and the max-lines ratchet all clean.

@nwparker gentle ping — happy to adjust anything.