## What / Why
This PR finishes the desktop refactor: the Tauri app is now a thin shell that launches the web server as a sidecar and loads the UI from `http://127.0.0.1:<port>`. All real backend logic lives in `packages/web/server/index.js`; desktop Rust keeps only native integrations (menu/dialog/notifications/updater/deep-link + window chrome).
This unblocks:
- consistent behavior across web/desktop/vscode (single backend)
- simpler desktop maintenance (no duplicated Rust backend)
- host switching between Local + remote instances in desktop
- reliable cold-start behavior on slow machines (VSCode + desktop)
## Key changes
- Desktop sidecar runtime
- build pipeline to bundle web dist + `openchamber-server` sidecar (`packages/desktop/scripts/build-sidecar.mjs`)
- robust local port selection (prefer saved/default, fallback to random; persisted in `~/.config/openchamber/settings.json`)
- improved PATH handling so the sidecar can locate `opencode` CLI (incl `~/.opencode/bin`, overrides, common bins)
- disable native right-click context menu in production builds (dev keeps it)
- Desktop instance switcher (Tauri-only)
- header button + modal to add/edit/delete remote hosts, set default, probe status/ping, switch back to Local escape hatch
- auth gate includes host switcher so you can recover when a remote host is broken/auth-required
- host list stored desktop-locally (not tied to the currently selected remote server)
- Notifications
- decision logic moved server-side; desktop notifications emitted via sidecar stdout and shown natively by Tauri
- prevent double-notifications on desktop Local origin (UI ignores SSE notification when native path is active)
- restore macOS notification sound
- Updates
- Tauri updater used only when viewing Local instance in desktop shell (avoid “remote web update” triggering desktop restart)
- Settings persistence & UX polish
- persist model favorites/recents via `/api/config/settings` (works for web + desktop; not origin-dependent)
- persist per-project sidebar collapse state in `projects[].sidebarCollapsed` via `/api/config/settings` (with debounce on toggles)
- macOS header sizing/traffic-lights offsets fixed (marketing macOS major injected from desktop; MultiRun header aligned)
- VSCode cold-start: keep retrying provider/agent loads after connection to avoid empty UI on slow machines
- misc lint/type fixes + bun.lock sync
- Desktop bootstrap / resiliency
- show onboarding screen when OpenCode CLI is missing (desktop Local origin), with retry hook to restart OpenCode after install
## Testing notes
- Desktop (macOS): switch Local <-> remote, set default host, verify auth gate recovery, native notifications (with sound), updater gated to Local
- Web: favorites/recents + per-project collapsed state persist across reload/restart
- VSCode: slow startup no longer results in missing providers/agents/models
* feat: extend quota providers and add usage dropdown
Add new quota providers:
- Claude (Anthropic API)
- Codex (OpenAI/ChatGPT)
- GitHub Copilot (base + add-on)
- Kimi for Coding
- OpenRouter
UI enhancements:
- Add rate limits dropdown in header with timer icon
- Desktop: sticky header with Used/Remaining toggle, refresh button
- Mobile: full-screen dropdown with same controls
- Per-provider sticky headers with provider logos
- Auto-refresh on open if no data exists
Usage settings page:
- Provider icon next to header
- Show in dropdown toggle per provider
- Global display selector (Usage vs Quota remaining)
- Auto-refresh controls in sidebar
Settings persistence:
- Add usageDisplayMode and usageDropdownProviders settings
- Server-side sanitization for new settings
Provider logo aliases:
- codex -> openai, claude -> anthropic
Bug fixes:
- Fix React error #310 by calling hooks before early returns
- Add aria-describedby to SettingsWindow dialog
- Remove horizontal scroll from dropdown
* fix: Remove duplicate github copilot declarations
* refactor(assets): rename logo svg files to include dimensions
* feat(assets): add 192x192 png versions of logos for PWA support
* fix(pwa): update static webmanifest with new logo paths and sizes
* fix(pwa): update dynamic manifest generation to use new 512px and 192px icons
* types: add optional context parameter to generatePullRequestDescription signature
* api: update gitApi to pass optional context to PR description generation
* http: include trimmed context in PR description API request body
* ui: add additional context input for PR generation with desktop disclosure and mobile sheet
* desktop: pass optional context to Tauri generate_pr_description command
* tauri: add optional context parameter and inject into PR description prompt
* server: read optional context from request and inject into PR description prompt
* vscode: update type signature for context parameter (compatibility only)
* fix(vscode): drop context from pr-description payload
Remove context field from PR description payload
Send only base and head to the bridge API for PR descriptions
Clarify payload compatibility across web/desktop environments
- Implement rate limiting for login attempts (10 attempts per 5min window, 15min lockout)
- Add Retry-After header and 429 response for rate-limited requests
- UI shows rate-limited state with countdown message
- Separate limits for identified IPs vs unknown clients (3 attempts without IP)
- Add cleanup mechanism for stale rate limit records
Co-authored-by: Jovines <jovines@qq.com>
* feat(openchamber): persist usage auto-refresh settings
Enable a switch to toggle automatic usage refresh
Provide input to configure refresh interval in milliseconds
Persist changes to desktop settings and server API when changed
* feat: add UsageCard component
Add a new UsageCard component to display a usage window with title, optional subtitle, and a progress bar
Show current usage percentage and a reset time label for the window
Render a formatted window label and a compact subtitle for concise UI
* feat(usage): add UsagePage UI for quota usage
Add a dedicated UsagePage with provider-based usage details
Show last updated time and auto-refresh status
Handle empty, not configured, and error states with informative banners
* feat(usage): add UsageProgressBar component
Introduce a new UsageProgressBar component to visualize quota usage
Display a gradient fill that changes with critical, warn, or normal tones
Expose accessible progress attributes for screen readers
* feat(usage): add UsageSidebar component
Display quotas for all providers in a scrollable sidebar
Refresh quotas with a button and loading indicator
Colorize provider rows based on usage status and runtime context
* feat: add usage section to Settings
Add a new Usage item to the Settings sidebar for desktop and mobile
Render UsagePage when the Usage tab is selected in Settings
Wire up new UsageSidebar and UsagePage components under usage
* feat: add Usage section to sidebar
Add new Usage section in the sidebar for API quota monitoring.
Display a bar chart icon and description for the Usage item.
Monitor and display API quota usage across providers.
* feat(desktop): add usage auto-refresh settings
Enable automatic refresh for usage data with new settings
Store refresh interval in milliseconds for usage updates
* feat(persistence): support usageAutoRefresh and usageRefreshIntervalMs
Persist usageAutoRefresh in desktop settings
Persist usageRefreshIntervalMs in desktop settings
Validate types for new fields during sanitizeWebSettings
* feat(quota): export providers and utilities
Expose QUOTA_PROVIDERS and QUOTA_PROVIDER_MAP for consumers
Export QuotaProviderMeta type for user code
Make formatting and usage resolution utilities available from quota module
* feat(quota): define base quota provider interface
Define QuotaProvider interface with id, name, isConfigured, and fetchQuota
Expose ProviderResult type in fetchQuota contract
* feat: add quota providers index and map
Expose QUOTA_PROVIDERS with OpenAI, Google and z.ai
Provide QUOTA_PROVIDER_MAP for quick provider lookup by id
* feat: add quota utils for percent formatting and tone
Add clampPercent to sanitize and clamp numbers to 0-100
Add formatPercent to render '-' for null and 'x%' for values
Add resolveUsageTone to categorize percent as safe, warn, or critical
* feat: add useQuotaStore for quota data
Load usage settings from desktop, VSCode, or API at startup
Fetch quotas for all providers in parallel and update loading state
Expose lastUpdated timestamp and error state for UI feedback
* feat: export quota types from quota module
Expose quota-related types in UI type definitions
Allow downstream code to import QuotaProviderId and related types
Aggregate quota exports under the quota module in index
* feat: add quota types for usage providers
Add QuotaProviderId and UsageWindow shapes to model quota data
Add ProviderUsage, ProviderResult, and related usage mapping for providers
* feat: add quota provider endpoints API
List available quota providers via GET /api/quota/providers
Retrieve quota details for a specific provider with GET /api/quota/:providerId
Log errors and return 500 with error message on quota fetch failures
* feat: add quota providers discovery and formatting
Detect configured quota providers from auth and account files
Normalize auth entries to tokens or objects for API usage
Expose formatted reset times and remaining window metrics
* feat: persist usage settings in UsageSidebar and remove from defaults
Load usage settings on mount for the sidebar
Persist changes to auto-refresh and refresh interval to server
Remove usage settings state and effects from DefaultsSettings
* fix(usage): guard auto-select in UsagePage when results empty
Guard auto-select in UsagePage when results are empty
Prevent unexpected provider selection on initial render
* fix(quota): set error state during quota updates
Reset error to null when a new quota result is added
Set error to the error message on fetch failure or fallback
Keep error state alongside results in all update paths
* feat: added themes system
* feat: smart sidebar auto-hide for files/diff tabs + lower files sidebar threshold
* feat: added Checkbox component and update theming
- Add reusable Checkbox component for toggles across UI
- Replace several inputs with Checkbox in settings and commands panels
- Add DiffIcon and apply surface/border theming to key UI areas
* feat: Add convert-vscode-theme.cjs to convert VS Code themes to OpenChamber format
* refactor: remove unused permission logic from ChatInput
- Remove unused permission rules parsing logic from ChatInput
- Memoize renderTheme in DiffWorkerProvider to avoid unnecessary recalculations
- Remove forceOpaque helper in vscode theme adapter
* fix: guard VSCode theme loading in MarkdownRenderer
* feat: add custom user themes loading and reload
- Load user themes from ~/.config/openchamber/themes at runtime
- Expose /api/config/themes to fetch custom themes
- Allow theme reloading from Settings → Theme → Reload themes in the UI
* feat: add notifyOnSubtasks setting to control subtask notifications
- Add notifyOnSubtasks toggle in NotificationSettings (moved to general section)
- Update useEventStream.ts to skip notifications for subtasks when disabled
- Add server-side push notification control for subtasks
- Update UI store, persistence, and desktop settings types
* feat: apply notifyOnSubtasks across web push + desktop notifications
---------
Co-authored-by: Jovines <jovines@qq.com>
* chore: upgrade @opencode-ai/sdk to 1.1.40
* chore(ui): bump @opencode-ai/sdk to 1.1.40
* feat(chat): add mobile controls drawer and panel switching
Add mobile-only controls drawer with open/close and panel switching
Introduce state and handlers for mobile controls and panels
Reset mobile UI state when switching to non-mobile or returning to unified controls
* feat: add mobile chat controls utilities
Add utilities to compute and display the selected agent and model names in mobile chat controls.
Add effort variant formatting, serialization, parsing, and ranking helpers for quick options.
Expose helpers to build quick effort option lists for the UI
* feat(ModelControls): allow external mobile panel control
Initialize mobile panel state from external props when provided
Fall back to internal mobile panel state when external control is absent
Notify parent on panel changes via onMobilePanelChange callback
* feat(chat): add StatusChip component
Add StatusChip button that displays agent, model, and effort
Show marquee animation when the label is truncated
Bind to config and session stores to reflect current context
* feat: add UnifiedControlsDrawer chat controls
Add a side panel to switch agent, model, and effort quickly
Show recent agents and models for faster reselect
Persist agent/model/variant selections in session
* fix(openchamber): correct base64 to Uint8Array typing
* feat: add reduced-motion support for marquee animations
Add marquee-text--auto to enable continuous scrolling
Introduce prefers-reduced-motion media query to disable animations
Apply reduced-motion rules to active marquee and hover states
* feat[worktrees]: enhance sdk worktree removal with fallbacks
Add dynamic resolution of remove/delete/archive methods for worktrees
Fallback to delete or archive when remove is not available
Throw clear error when SDK version does not support worktree removal
* feat(ui): track recent agents and efforts in UI store
Add recentAgents array to UI state for quick access
Track up to 5 variants per provider/model in recentEfforts
Expose addRecentAgent and addRecentEffort actions to update history
* fix(deps): upgrade @opencode-ai/sdk to 1.1.42
Upgrade the OpenCode AI SDK to 1.1.42 across packages
Refresh lockfile entries to reflect the new SDK version and integrity hash
Ensure downstream packages consume the latest SDK and remain compatible
* refactor: simplify agent overflow logic in UnifiedControlsDrawer
- Update startGlobalEventWatcher to track session activity phases independently of UI, mirroring Tauri desktop behavior.
- The web server now maintains an always-up-to-date activity cache by listening to SSE events continuously, even when the UI is hidden.
Add SDK-based worktree management that lists and starts SDK worktrees
Migrate per-project setup to ~/.config/openchamber/<projectId>.json
Deprecate .openchamber legacy paths and adapt UI to new config
Add paging support when loading ClawdHub skills
Retry on API errors and throttle for ClawdHub requests
Load source content on source change and show loading indicators
* feat: display provider logos for favorite/recent models
Show provider logo next to model name in favorites and recents
Render provider logos in ModelControls, ModelMultiSelect, and ModelSelector lists
Maintain zero-logo state for other sections to avoid clutter
* feat: render user message as markdown instead of plain text
Render agent mentions as markdown links in user text
Apply inside list style for chat content to fix list rendering
Rely on SimpleMarkdownRenderer for consistent rendering
* fix(openchamber): adjust layout and overscroll behavior
Enable overscroll-auto on overlay containers for smoother scrolling
Move page content to full-width wrapper and preserve section borders
Show AboutSettings inside its own bordered block when visible
* feat: integrate GitHub auth status store and UI
Introduce GitHubAuthStore to track connection status and polling
Show GitHub avatar in header when connected
Guard issue/pr dialogs behind GitHub auth status and show notices
* feat: add GitHub multi-account support
Add API and UI flow to activate a GitHub account
Show and switch between multiple GitHub accounts in header
Persist and normalize accounts list with current selection
Introduce a hook to detect text truncation for dynamic UI updates
Replace static marquee spans with a reusable marquee component in files and labels
Enable copying of terminal selection on mouse up and touch end
* feat: add chat virtualization and turn grouping infrastructure
Implement VirtualMessageList with virtualization and overscan for smooth scrolling
Refactor MessageList to render only visible messages and pass scroll refs
Introduce TurnGroupingContext and provider to stabilize per-turn rendering and reduce re-renders
* feat: add web server session activity endpoint for visibility restore
Add /api/session-activity endpoint to expose tracked session activity
Use web server activity first when restoring visibility, with fallback to global status
Update server SSE to set session phase on activity events
* feat(chat): split TurnGroupingContext into UI/Streaming contexts
Add separate UI state and streaming contexts to reduce re-renders.
Skip animations for items visible in collapsed view when expanding.
Track shown parts in collapsed mode to fade in progressively
* feat(ui): cache turn groups with expand state and guard web activity
Add isExpanded to the turn cache key to trigger updates correctly
Expose isGroupExpanded from UI state so groups reflect expand/collapse
Limit web server session activity fetch to web runtime only
* fix(server): improve external server and error handling
- Skip OpenCode shutdown when using external server
(OPENCODE_SKIP_START=true)
- Return 404 status for non-existent directories instead of
empty array
- Move error logging before response handling
* feat(ui): add configurable text justification activity setting
- Replace hardcoded ENABLE_TEXT_JUSTIFICATION_ACTIVITY constant with
dynamic showTextJustificationActivity setting in useUIStore
- Add toggle UI in Chat settings section (Settings > Chat)
- Add setting to DesktopSettings type for cross-runtime persistence
- Default value is false (matching original constant behavior)
---------
Co-authored-by: iiyangdianfeng <goodydfwow@gmail.com>
* feat: add plan/build mode switching and Plan view tab
Add PlanView view and integrate plan tab into main layout
Introduce plan_enter and plan_exit tools with icons and status strings
Persist per-session agent selections in context to reduce UI flicker during mode switches
* feat: enchanced plan discovery checks in header and plan view
* fix(ui): align PlanView and Header with sessionDirectory logic
Remove dependency on context store in header and compute sessionDirectory from session or currentDirectory
Compute display and repo paths using sessionDirectory for PlanView and update path resolution
Poll plan content every 3 seconds to reflect changes without heavy retries
* feat(ui): improve header tab switching and add copy buttons for plan
Switch header tab navigation to central UI store and auto-switch when plan is unavailable
Show checkmark icon after copying file contents or path with auto-hide timeout
Extend plan view path resolution to prefer session directory when present
Add check details to PR context via includeCheckDetails flag
Open a checks dialog showing check run summaries and steps
Improve PR lookup for forked repos by matching head branch
Add atomic write for the shared storage file to prevent race conditions.
Write to a temp file with restricted permissions and atomically rename.
Apply 600 permissions to the final storage file when possible
Add GitHubIssuePickerDialog UI for selecting issues
Enable new session from GitHub issue from session sidebar
Implement GitHub issues/list/get/comments APIs across desktop, web, and VS Code
* feat: integrate GitHub OAuth device flow across runtimes
Add GitHub OAuth device flow endpoints across runtimes
Introduce GitHubSettings UI panel and sidebar entry
Persist GitHub auth state in per-runtime storage
* feat: add GitHub PR status and PR description generation
Show PR status for the current branch in the Git view
Generate a pull request description from the diff between base and head
Expose prStatus, prCreate, and prMerge APIs in web and desktop clients
* feat: add GitHub PR ready for review
Add API to mark pull requests as ready for review
Show a Ready button for draft PRs and reflect status in UI
Handle token expiration and GraphQL errors when marking ready
gsxdsm