Adds a self-contained <CommandPalette /> client component plus a
static data file that lists every admin page (per the new IA in
docs/superpowers/specs/2026-06-17-admin-redesign.md §4) and a
small set of quick actions.
Behavior:
- Toggles on Cmd+K (mac) / Ctrl+K (win/linux) at document level.
- Escape or backdrop click closes; Enter or click navigates.
- Fuzzy case-insensitive match on label / category / keywords,
top 8 results, highlighted match substring.
- ↑/↓ keyboard nav, mouse hover updates selection, auto-focus
on input when opened, body scroll lock while open.
- Visual: uses existing --admin-card-bg / --admin-border /
--admin-primary-soft / --admin-radius-lg / --admin-shadow-lg
tokens; fade-in 180ms, scale 0.98→1 animation.
- Returns null when closed (no DOM noise).
Out of scope for this pass:
- Component is NOT mounted in the admin layout (main thread wires
it in). The data file is the source of truth; the palette does
not import from AdminSidebar.
- localStorage-based Recent items are left as a TODO comment
in the component; v1 ships without persistence.
No new dependencies; uses lucide-react (already in package.json).
npx tsc --noEmit is clean.
- --admin-bg: warm cream (#FAF7F0)
- --admin-text-primary: ink black (#1A1814)
- --admin-text-muted: warm grey (#8A867E) — was too dim
- --admin-border: soft beige (#E8E4D7) — was too dirty
- --admin-primary: deep botanical (#1F4D2A)
- --admin-accent: warm amber (#B8761E) — the 'one thing to do'
- --admin-warning: amber (was #aba278 — looked like dirt)
- --admin-danger: warmer rust (#A8321C)
- sidebar: deeper, more refined (#2A2520)
- drop purple/blue/citrus stat icons in DashboardClient; use semantic tokens
- use --admin-warning instead of hardcoded #f59e0b in usage bar
- design spec: aesthetic, color tokens, IA, phased plan
- checklist: 7 phases, every task has notes column, revert cheatsheet
- working branch: design/ui-revamp-2026-06
- Remove Google Fonts @import from 4 components (SiteHeader, LandingPageWrapper,
TestimonialsAndCTA, FeaturesAndStats) — eliminate render-blocking external
request and font flash
- Replace all unloaded Cormorant Garamond / Playfair Display / DM Sans /
Plus Jakarta Sans references with the existing next/font variables
(Fraunces, Manrope, Fragment_Mono) — visual coherence with the design system
- Update 9 pages (blog, brands, changelog, maintenance, protected-example,
roadmap, security, waitlist, WaitlistForm) to use the loaded Fraunces
- Fix dead --font-geist / --font-jetbrains-mono references in
admin-design-system.css (now point to --font-manrope / --font-fragment-mono
which are actually loaded)
- Add atelier-pill, atelier-numerals, atelier-fineprint, atelier-canvas-soft
utility classes; .atelier-input:focus-visible refined ring
- Add .ha-field-textarea, .ha-scroll, .ha-skeleton to admin design system
- Extend prefers-reduced-motion guard to atelier-enter/stagger/shimmer
- Apply atelier-fineprint to login/not-found/error pages for consistency
- No structural changes; build passes, 0 type errors
User pain point: skeleton loading.tsx files made the app feel like
a sequence of page reloads, exposing backend latency. Replaced with
a single 1px shimmer bar + crossfade via React's <ViewTransition>.
Changes:
- Enable experimental.viewTransition in next.config.ts
- Add SmoothViewTransition wrapper (ViewTransition name=page-content)
- Add LoadingFade component: thin animated bar instead of skeleton
- Add RouteAnnouncer for a11y (screen readers + focus reset)
- Add ::view-transition-old/new CSS for the crossfade (220ms, no
jarring slide, respects prefers-reduced-motion)
- Wrap admin/tuxedo/IRD layout children in SmoothViewTransition
(sidebar/header/footer stay mounted; only the body fades)
- Replace 19 skeleton loading.tsx files with the fade component
Result: navigation now feels like a single app, not a series of
preload-and-render events. The user never sees a 'skeleton of the
page they're about to load.'
Visualizes the current layout nesting in the App Router, identifies
gaps where shared sub-layouts could be added (cart/checkout, wholesale,
water, admin/communications, admin/settings), and shows how to add
a new nested layout or parallel route for modal-style navigation.
- Use StopTableClient in server page component
- Add view mode toggle (table/card) matching products page
- Match stats cards styling to products page
- Add consistent filter bar layout with AdminViewModeTabs
- Add card view for stops with inline delete confirm
- Improve row actions with Edit button + dropdown menu
- Add 'Today' badge and past stop dimming in card view
- Add EditStopModal for unified add/edit stop workflow
- Enhance StopTableClient with sortable columns (date, city, location, status)
- Add stats bar showing total, active, upcoming, and draft counts
- Improve pagination with 'Showing X-Y of Z' indicator
- Add 'Today' badge for current-day stops
- Dim past stops visually
- Add Edit action to row menu with icons
- Reduce page size to 25 for better UX
- AdminSidebar now renders BrandSelector when brands are available
- Stops page uses getActiveBrandId() instead of adminUser.brand_id
directly, so platform admins see all stops and brand selection
via the sidebar picker works correctly
- BrandSelector already existed and was wired up in the layout,
just never rendered in the sidebar
The page was importing from @/lib/supabase which is a mock client that
returns empty results. Replaced with pool.query() against the real
Postgres stops table.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The seed step on every deploy was inserting duplicate stops each run.
The cleanup script (scripts/cleanup-duplicate-stops.ts) can be used to
dedupe existing rows when needed — run manually via:
DATABASE_URL="..." npx tsx scripts/cleanup-duplicate-stops.ts
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
psql with Neon pooler times out on large batch RPCs. The tsx script
replaces -pooler. with direct compute endpoint and uses pg Pool with
proper batching + progress output.
Run db/seeds/2026-tuxedo-tour-stops.sql via psql after migrations in
the deploy step. Uses admin_create_locations_batch + admin_create_stops_batch
RPCs (migration 0003) to insert 40 locations + 269 stops for the Tuxedo Corn
2026 tour. Wrapped in BEGIN/COMMIT with DELETE-first semantics — idempotent.
- Fetch dashboard stats server-side in admin/page.tsx to eliminate
client-side useEffect waterfall and the '—' placeholder flash
- Pass pre-fetched stats as props to DashboardClient
- Lazy-load UpgradePlanModal via next/dynamic (only loads when needed)
- Redesign stats cards with compact layout, smaller icons, Fraunces serif values
- Improve Quick Actions + Usage row: side-by-side on desktop, stacked mobile
- Clean up Recent Orders as a proper list with icon/name/time/amount/badge
- Add staggered fade-up entrance animations (0/60/120/180/240ms delays)
- Consolidate loading.tsx skeleton to match actual dashboard structure
- Mobile-responsive: 2-col stats on mobile, stacked usage, collapsible header
- LandingPageWrapper: Use dynamic year for copyright
- HeroSection: Add float-slow/float-slow-delayed keyframes, add prefers-reduced-motion media query
- TuxedoVideoHero: Add reduced-motion JS guard and CSS media query for accessibility
The inline node -e '...' blocks inside the YAML run: block had shell
quoting conflicts (nested parens, single-quote escapes) that caused
act/local tooling to fail with 'syntax error near unexpected token ('.
Split into two dedicated scripts:
- scripts/preflight-check.js — neon_auth schema check + 0001_init.sql tracking repair
- scripts/postflight-check.js — admin_users table verification after migrations
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The "Run migrations" job (and thus the whole deploy) was failing on every push after the first successful bootstrap with:
✗ 0001_init.sql failed: relation "admin_users" already exists
Root causes:
- 0001_init.sql used plain CREATE TABLE/INDEX/TRIGGER despite the header comment claiming "CREATE TABLE IF NOT EXISTS".
- The _migrations tracking row for 0001_init.sql was never recorded on the prod DATABASE_URL (tracking logic landed after initial apply, or apply happened outside the runner).
Fix (3 layers of defense):
* db/migrations/0001_init.sql: every CREATE TABLE now uses IF NOT EXISTS (63 tables), CREATE INDEX uses IF NOT EXISTS (49), all CREATE TRIGGER wrapped in safe DO $$ IF NOT EXISTS (pg_trigger join check) THEN ... END IF; $$, removed the file's own BEGIN/COMMIT so the runner's transaction is authoritative. RLS drop+create and CREATE OR REPLACE FUNCTION were already safe.
* scripts/migrate.js: added ensureTracked() right after loading the applied set. For 0001/0002, if the core objects (admin_users table, password_hash column) already exist in information_schema but the tracking row is absent, INSERT the filename (ON CONFLICT DO NOTHING) and log "repaired tracking". Then the normal skip path handles it.
* .gitea/workflows/deploy.yml: added an inline node -e pre-repair (same admin_users existence check → force _migrations row) immediately before `npm run migrate:one`. This protects even on an older checkout of the runner. The neon_auth preflight and post-apply "admin_users must exist" hard gate are untouched.
* Minor: cleaned 0002_admin_password.sql tx wrapper for consistency; updated MEMORY.md with the incident + resolution.
Result: deploys (and any server-side `node scripts/migrate.js` recovery runs) on an already-initialized prod DB will repair/skip 0001 cleanly, pass the verification query, and continue to build/deploy. The scp of scripts/ + db/migrations/ in the Deploy step now carries the fixed versions.
See the plan note in deploy.yml and MEMORY.md for background.
- getAdminUser now properly supports platform_admin with 0 brand links and loads full brand_ids
- createAdminUser action now inserts into admin_user_brands join table
- Admin layout surfaces the signed-in email on Access Denied
- AdminAccessDenied links to /login instead of dead-end /admin
- Main dashboard uses direct pool query instead of dead supabase shim
- Improved provision-admin.ts script for prod bootstrap (loads .env.production too)
The membership query in getAdminUser() was selecting role from
adminUserBrands.adminUserId, which is wrong - admin_user_brands has no
role column. Role is stored in admin_users. Also added try/catch
around withPlatformAdmin to prevent DB errors from throwing.
- Add MinIO/S3-compatible storage client (src/lib/storage.ts) with uploadObject,
deleteObject, presigned URL helpers, and BUCKETS constant
- Wire product images, brand logos, and water log photos to MinIO via the
new storage client
- Migrate forgot-password to Neon Auth (remove Supabase /auth/v1/recover call)
- Migrate send-scheduled cron to direct Postgres + Resend (remove Supabase Edge
Function proxy)
- Add logoUrl to email types (OrderReceipt, Welcome, PasswordReset) and pass
brand_settings.logo_url from all call sites
- Update email templates to use dynamic logoUrl instead of hardcoded Supabase
bucket URLs
- Remove hardcoded Supabase URLs from TuxedoVideoHero, TuxedoAboutPage,
TimeTrackingFieldClient; use brand_settings props + local public/ fallback
- Download brand logos (3) and tuxedo-hero.mp4 (36MB) from Supabase bucket to
public/ for local development
- Add MinIO env vars to .env.example (endpoint, access key, secret, buckets)
- Fix TimeTrackingFieldClient to destructure logoUrl and brandAccent props
- Fix admin/users.ts logoUrl type (null → undefined for optional string)
- Remove stale sb- cookie from wholesale-auth
- Migrate tuxedo/about page to remove supabase import and use pool query for
wholesale_settings lookup
The products list page at /admin/products still queried the legacy Supabase
mock client, which returns an empty array. Combined with the SaaS schema
renaming brands → tenants, the platform admin's brand picker for new
products was silently failing (getBrands SELECTed from a non-existent
'brands' table), making 'Add Product to catalog' look like a no-op.
- src/app/admin/products/page.tsx: replace supabase.from('products') with
a Drizzle query using withPlatformAdmin (cross-tenant) or
withTenant(brandId, ...) (scoped). Map new columns back to the legacy
Product shape the existing UI consumes:
* price_cents (integer) → price (dollars)
* tenant_id → brand_id
* product_images LEFT JOIN for first image per product
* default type='pickup', is_taxable=false (columns not in SaaS schema)
- src/actions/admin/users.ts (getBrands): query tenants table instead of
the legacy brands table.
Verified locally against the seeded dev DB: 8 products across 2 tenants
now render in the catalog, and the brand picker populates with both
tenants for platform admins.
Previously, any non-empty AUTH_GOOGLE_ID + AUTH_GOOGLE_SECRET would
trigger the Google button. That broke dev/CI setups where the env
vars are placeholder strings (e.g. 'dummy-google-client-id') — the
button would render and immediately 401 from Google with 'invalid_client'.
Real Google OAuth client IDs always end in '.apps.googleusercontent.com'.
Gate hasGoogle on that suffix so the login page falls back to the
credentials form (or the 'not configured' message) when the values
aren't real.
The seed script previously imported hashPassword from src/lib/passwords.ts,
which has 'import "server-only"' at the top. That guard throws when the
file is loaded outside a Next.js server runtime (which is what 'tsx db/seed.ts'
is — plain Node). Inline the same scrypt format (matches verifyPassword)
so the seed runs from a CLI.
Gitea deploy failed with:
Module not found: Can't resolve '@stripe/stripe-js'
Module not found: Can't resolve '@stripe/react-stripe-js'
These were imported by src/components/storefront/StripeExpressCheckout.tsx
(transitive: src/lib/stripe-client.ts → src/app/checkout/CheckoutClient.tsx)
but never declared in package.json. Add them so the Gitea build can resolve
the imports.
Tyler
openclaw
tyler