Returning to the page via browser Back (e.g. from the /download
redirect's external target) restored it from the bfcache in Firefox:
the JS heap thaws but the WebGL context is discarded, so the galaxy
canvas came back blank and the rAF-driven instruments looked
unmounted. Chrome restores GPU state, which is why it never showed.
The pre-paint head script now attaches a pageshow listener that
reloads on e.persisted — the listener lives inside the frozen page,
so it fires on the restore itself. The intro is seen-flagged by
then and the page is statically prerendered, so the reload is cheap.
Claude-Session: https://claude.ai/code/session_01KpbGFYxEXuaYU9J75vTK9j
The per-arm conveyor offset was in absolute cycle units, so it
resonated whenever it landed near a multiple of the slot spacing:
at 51 cards (9 slots/arm, spacing .111) the .13 offset was one slot
+ 17%, radially aligning every arm's ladder into six-fold rings —
the starfish. 64 cards (spacing .0909) just happened to dodge it.
The stagger is now measured in slot gaps (divided by slotsPerArm), so
it can never ring-align at any card count or viewport scale, and the
default moves to 0.38 — the golden-ratio conjugate — which lands the
six arms at low-discrepancy phases within the gap.
Claude-Session: https://claude.ai/code/session_01KpbGFYxEXuaYU9J75vTK9j
With 17 clips on the CDN, 51 cards lands the round-robin deal at
exactly three cards per clip (down from ~4.6 repeats at 64). The
neighbor-gap collision sizing grows the cards to keep the same screen
fill, and smaller viewports still scale the count down from the same
Tuned @ Mpx reference.
Claude-Session: https://claude.ai/code/session_01KpbGFYxEXuaYU9J75vTK9j
The CDN holds 17 clips but the wall only dealt the first 14. Verified
015-017 return 200 (018+ 404). At the tuned 64-card count this drops
simultaneous repeats per clip from ~4.6 to ~3.8; the least-recently-
shown rebirth reassignment spreads what remains.
Claude-Session: https://claude.ai/code/session_01KpbGFYxEXuaYU9J75vTK9j
The max-width content container carried the site's border-x rails, so
two vertical hairlines cut through the immersive spiral/wordmark stage.
The poster now runs edge to edge; the structural rail-and-tick language
begins at the announce strip (the fold line), which keeps its border-x.
Claude-Session: https://claude.ai/code/session_01KpbGFYxEXuaYU9J75vTK9j
The #1916 restyling wrote `<LoaderCircleIcon spin>` in five spots, but
raw lucide-react icons have no `spin` prop — the attribute fell through
to the SVG (React DOM warning) and the spinner never animated. The two
spots inside ImagePromptRow were corrected during the services/main
merge resolution; this covers the remaining three files with the same
animate-spin class main uses. Also drops the unused DynamicIcon import
in create-video (pre-existing on both lineages).
`tsc --noEmit -p apps/artcraft-webapp/tsconfig.app.json` is now clean.
Claude-Session: https://claude.ai/code/session_01KpbGFYxEXuaYU9J75vTK9j
The design branch forked from the old monorepo main at f3442f3146;
main then advanced through #1910 (last-viewed badge), #1911 (iOS
upload fix — uploading queue stores previewUrl, not File), #1912,
#1699 (keybinds), the auth-page video update (976f2ae757, Vimeo
iframe → muted <video>), and the modal freeze fix (ed57c6de47,
auto-merged) before the artcraft/artcraft-services split.
Seven files conflicted where main's functional work crossed the
branch's #1913/#1916 webapp restyling. Resolution rule: main's
functionality, the design branch's styling —
- auth-showcase: main's <video> element, squared corners kept
- ImagePromptRow (webapp + libs): main's previewUrl refactor; kept
square corners/hairline borders. The branch's `spin` prop was a
latent bug (raw lucide icons have no such prop — no animation and
a React DOM warning), so main's animate-spin class wins.
- GalleryCard/GalleryRow: main's lastViewed ring + badge woven into
the branch's squared-corner classes (badge pill squared to match)
- old-site app.tsx: union of both new routes (landingp0 + creator)
- TimelineClipRow: main's enableClipTransition undo-step flow
superseded the branch's setClipTransitionEasing(DEFAULT_EASING)
artcraft-website-next is untouched by the incoming side. Webapp
typechecks clean after the follow-up spinner fixes.
Claude-Session: https://claude.ai/code/session_01KpbGFYxEXuaYU9J75vTK9j
Proofing-pass deployment plumbing. The app referenced seven marketing
routes it doesn't serve yet (/download, /pricing, /tutorials, /news,
/faq, /press-kit, /support) — every one now 307s to the shipping Vite
site, which this app already depends on staying live (MEDIA_BASE
feature footage). Drop a route from LEGACY_ROUTES as its page migrates.
netlify.toml follows the sibling apps' discovery pattern with base
frontend/apps/artcraft-website-next (standalone lockfile, not the Nx
workspace) and the Next runtime plugin, which also carries the
/cdn-media and /ext-media proxy rewrites into production.
turbopack.root and outputFileTracingRoot are pinned to the app dir —
Next was guessing the Nx workspace's lockfile as the project root,
which breaks output file tracing on serverless deploys.
Claude-Session: https://claude.ai/code/session_01YBoaBEt3gynzTwogBFtn4B
Adds a new creator spotlight landing page to the artcraft-website under
`/creators/`.
### Changes
- New spotlight page: hero with fanned reel triptych, credits marquee,
work grid, craft section, and closing CTA, all in the existing
landing-page design language
- Reel cards use locally hosted thumbnails (`public/images/creators/`)
and link out to the posts on Instagram, no embeds
- Primary "Create Now" CTA routes subscribers to the webapp and everyone
else to `/pricing`
- SEO wired up: `PAGE_META` entry for the prerender step, `_redirects`
rules plus short 301 aliases for the route
Recreate restored reference videos/audios from the prompt record with duration: 0. The cost quote forwards that as a zero input-duration hint, and the backend deliberately prices a zero/unknown input at the worst-case 30-second billing maximum (Seedance 2.5 bills reference-video input seconds), so a recreated generation quoted far above the original
charge, e.g. a 1080p 10s output with ~5s of input quotes 548 credits, but 1096 with the failsafe.
Measure each restored reference's duration from its CDN copy while building the recreate payload, the same way the library "Send to prompt" path probes durations. Unmeasurable files keep 0 (worst-case quote as before; generation always bills from its own server-side measurement).
Add Billing tab to webapp settings
- New "Billing" tab in the settings modal, between Account and API Keys
- Shows your current plan (Free/Basic/Pro/Max)
- Subscribers get a "Manage Plan" button that opens the Stripe portal,
same as the pricing page, plus a "View Plans" link for switching
- Free users get a "View Plans" button to the pricing page
- Handles signed-out and loading states like the other tabs
## Summary
Adds a unified, remappable keyboard-shortcut system
(`@storyteller/keybinds`) across the webapp and desktop app, replacing
the scattered hardcoded `keydown` listeners in the 3D editor, 2D editor,
and moodboard with one registry, two presets (**Gamer** = today's
defaults, **Blender-like**), a Settings → Keybinds pane, and a hold-Ctrl
/ pinnable cheatsheet on every editor surface.
This PR was approved in June and then sat behind ~195 commits on `main`
(timeline, build/record modes, lucide migration). It has been merged
with current `main` and extended to cover the features that landed
since. **Nothing changes by default** — the Gamer preset reproduces
every existing key exactly.
## Architecture
- **Registry of `ActionDef`s** keyed by surface-namespaced ids
(`pagescene.timeline.playPause`), bound to physical `event.code`.
Resolution: user override → preset delta → base bindings. Persisted in
the `artcraft-keybinds` zustand store.
- **Five surfaces**: `global`, `pagescene` (3D), `pagedraw` (2D),
`moodboard` (canvas), `moodboard-grid` (board library / inspector /
presentation).
- **Declared availability** — `ActionDef.when(ctx)` gates decide whether
a binding is live for the current editor state (`sceneMode`, `encoding`,
`timelineExpanded`, `timelineSelection`, …). An unavailable binding
falls through *without consuming the event*, which is what lets Space /
arrows / Delete / Escape be shared by context-exclusive actions. A
sampled-context checker (`actionsCoAvailable`) proves the exclusivity in
tests and drives the conflict UI in Settings.
- **Global surface is a pure trigger layer**: features register behavior
with `useGlobalAction(id, handler)`; a window capture-phase dispatcher
fires above the modal key-swallow, and an unregistered binding is inert
per host. First inhabitant: Ctrl/Cmd+B sidebar. This is the intended
home for the upcoming privacy-blur toggle (one `ActionDef` + one binding
+ one `useGlobalAction` call).
- Cross-surface conflict detection (global ↔ everything, both
directions); preset-integrity tests guard every preset against
same-surface collisions.
- **Video editor keeps its own keybinds system** — deliberate non-goal.
## What's in it
**3D editor (pagescene)**
- Timeline actions: Space play/pause, ←/→ frame step or keyframe/strip
nudge, Home/End, `I` add keyframe (Blender parity), Shift+T expand,
Delete/Backspace on a valid timeline selection (exact complement of
scene-object delete).
- Record mode: Tab toggles build/record, `C` capture still, `R` record
video (bare letters — Firefox eats Ctrl+Shift+S/E), Escape cancels an
in-flight encode (overlay Cancel button too). Record mode is immutable:
all scene-mutating actions gate off, Escape no longer drops out of the
render cam.
- **Per-edit timeline undo** — every timeline edit is one Ctrl+Z step
(`TimelineEditAction` with whole-timeline snapshots; drags and the
easing popover coalesce into single steps). Save now only moves the
Cancel baseline; Cancel is itself undoable.
- The old "Shortcuts" modal is replaced by the live cheatsheet, pinnable
from the same button.
- Dead code removed: `MouseControls.onkeydown`, `OnboardingHelper`, the
hardcoded `Help` view, three unused pagedraw hotkey hooks.
**2D editor (pagedraw)**
- Tool hotkeys 1–5 with live key badges on the toolbar icons; the shape
key cycles rectangle → circle → triangle (submenu-aware).
- Ctrl+= / Ctrl+- (and numpad) resize the active tool. The eraser now
has its own size (was sharing the brush's) and its own slider popout;
also fixes the eraser button never highlighting.
**Moodboard**
- Canvas keys unchanged (now registry-backed), plus Ctrl+0 reset
viewport.
- Grid: rate 0–5 / `P` pick / Delete / Escape, inspector and
presentation ←/→/Space/Esc — all rebindable. Escape now closes the
inspector; arrows no longer navigate while typing in the tag field.
**Settings & cheatsheet**
- Settings → Keybinds: preset picker, name/key search, per-action rebind
with conflict warnings, per-section reset, and a bottom "danger zone"
reset-all with a two-step confirm (2 s lockout + visible 5 s
auto-disarm).
- Cheatsheet: hold Ctrl (⌘) 3 s to peek; **General → "Keep shortcut
cheatsheet open"** makes it stick until Esc / click-outside; the 3D
"Shortcuts" button pins it.
## Testing
- `@storyteller/keybinds` unit tests (29): resolution layering,
overrides, per-surface/global reset, conflict detection incl.
global↔surface, matcher semantics, and preset integrity across all
presets.
- Builds green: `keybinds`, `ui-pagescene`, `ui-pagedraw`,
`ui-moodboard`, `ui-settings-modal`, `artcraft-webapp`, `artcraft`
(desktop).
- Manual QA pass completed on the webapp for every area above (settings
pane, cheatsheets, Blender modal transforms, timeline/record keys,
pagedraw toolbar, global surface, reset UX, pinned cheatsheet, timeline
undo, Escape contexts).
## Notes for review
- The keybinds store is unversioned-migration-free on purpose — the PR
was never merged, so there's no persisted shape to migrate from.
- Space-hold pan (moodboard) and held WASD camera movement stay as
gestures, resolved from the same store but outside the discrete-action
dispatch.
---------
Co-authored-by: bflatastic <bombayapirat@gmail.com>
Tuner (see TUNER.md): a draggable, collapsible dev panel rendering every
group registered via defineTunables() as slider + number-input rows.
- Live: physics/light/material/halo values are read per frame; formation
values trigger a debounced re-sample
- Values persist to localStorage; modified labels tint accent; Copy puts a
JSON snapshot on the clipboard for baking back into defaults; Reset clears
- Shown in dev builds or with ?tuner=1; drag position and section collapse
state persist
- Policy (documented): every new tweakable constant registers with the tuner
by default
- Hero groups: Formation, Physics, Cursor & Light, Material, Halo
Formation: the hex lattice read too uniform. Balls are now placed by
multi-pass dart throwing over the glyph mask — big balls claim spots first,
smaller ones settle into leftover gaps, random depth within the slab, 3D
overlap rejection via spatial hash — a faked settled-pile impression, not a
physics solve. Stroke-margin test keeps big balls inside the letterforms so
the type stays crisp.
Drag and drop didn't work while the prompt focus mode was open - the
modal sat on top of the inline promptbox's drop zone, so files just did
nothing (webapp and desktop).
### Changes
- `usePromptBoxDrop` now exposes `fullscreenDropZoneProps` - same
handlers/routing, second ref so the Tauri hit-test follows the
focus-mode modal while it's open
- `PromptFullscreenModal` (shared lib + webapp copy) takes optional
`dropZoneProps` / `dropOverlay` and renders the drop zone + hover
overlay on its panel
- Image / Video / Audio boxes + webapp `PromptBox` pass their existing
drop wiring through - one overlay element now serves both the inline box
and focus mode
- No new drop logic anywhere: same MIME routing, accept/reject toasts,
keyframe handling, and paste behavior as dropping on the inline box
On iOS, adding more than 2 image prompts can blow past the browser's
memory limit and force-reload the page, wiping your prompts. We were
storing every uploaded image as a full-res base64 string just to show a
56px thumbnail.
- Thumbnails are now small ~512px previews instead of full-res data URLs
(memory per image goes from ~100MB+ to under 1MB)
- Tapping a thumbnail still shows the original full-res image
- Nothing changes for the backend — uploads and generation still use the
original file + media token
- Fixed the flicker when an upload finishes: the spinner card now turns
into the image card in place, instead of a new card popping in next to
it
- Small cleanups along the way: leaked object URLs in the old prompt
row, and upload spinners that never actually spun
Applies to webapp (desktop + mobile), website, and the desktop app since
they share the upload code.
- Ball sizes now vary organically: smooth spatial value noise drifts size
clusters across the word, white noise roughens per ball; the neighbor cap
keeps every draw collision-safe
- Radial halo billboard tracks the cursor BEHIND the text plane, so balls
silhouette against the glow (additive in dark, soft accent wash in light)
- Cursor point light switched to physical inverse-square falloff (decay 2):
balls near the cursor are genuinely brighter; the light dims to 30% when
the pointer leaves so the resting state stays moody
- Dark mode is an eclipse rig: near-black lacquer balls (clearcoat physical
material), almost no front light, cool key + counter rim from behind, and
no environment map — scene.environment floods instanced materials at full
strength regardless of envMapIntensity, which was washing the whole word
to gray; light mode keeps the procedural room reflections
The hero brand mark is now the signature interactive moment: 'artcraft'
rasterized per-glyph (manual advances so no ligature shaping), sampled onto a
jittered close-packed lattice 2-3 layers deep, and rendered as instanced 3D
spheres under a perspective camera fitted so world units equal CSS pixels at
the text plane.
- Cursor is both a point light and a push force (radial + swirl, quadratic
falloff, toward-viewer pop); balls scatter, then rejoin on home springs
- Per-ball radii capped at half the nearest-neighbor gap, plus a 3D
spatial-hash collision pass, so spheres never clip through each other
- Mount plays an assemble: balls fly in from a scattered shell
- Idle: light wanders a slow Lissajous path; balls breathe subtly
- Fallbacks: static display-type wordmark for SSR/crawlers, reduced motion,
and WebGL failure; theme-aware colors via shared token watcher (extracted
to lib/theme-colors, now also used by the blocking/render comparator)
- Tagline 'Controllable AI for artists.' remains the crawlable h1
Standalone Next 16 app at frontend/apps/artcraft-website-next (excluded from
the npm workspace like artcraft-dashboard; React 19 + R3F 9 + GSAP + Lenis +
Tailwind 4). Branched from design/prototype0; the Marathon-style navbar is
ported to next/link with theme-aware tokens plus a light/dark toggle.
- Design system: hairline-grid 'instrument panel' language; light (blueprint
paper) + dark (viewport terminal) themes via CSS tokens, system-pref default
with persisted override applied pre-paint
- Landing: hero, capability ticker, feature grid (real product footage),
ownership, made-with (click-to-load YouTube), CTA, footer — all crawlable
SSR HTML; GSAP reveals are progressive enhancement only
- 3D hero: interactive blocking↔render comparator — wireframe R3F set clipped
over the product reel by a cursor-driven scan line; degrades to plain video
on reduced-motion/coarse-pointer/WebGL failure
- SEO: metadata + OG, sitemap.ts, robots.ts, SoftwareApplication JSON-LD
Removes Font Awesome entirely
### Changes
- All ~1,000 icon call sites now import Lucide components directly from
`lucide-react` (`<PlayIcon className="..." />`); no FA packages, tokens,
or `fa*` identifiers remain anywhere.
- New `@storyteller/icons` lib holds the pieces Lucide doesn't ship:
`DynamicIcon` for runtime-chosen icons (`icon={loading ?
LoaderCircleIcon : SaveIcon}`), 11 brand logos (Discord, TikTok, ...) as
inlined CC BY 4.0 paths, and HD/SD quality badges drawn in Lucide's
stroke style.
- Global `.lucide { display: inline-block; width/height: 1.2em }` rule
in each app stylesheet: keeps icons font-relative like FA (1.2em
compensates Lucide's 24px-viewBox/20px-grid padding) and undoes Tailwind
preflight's `svg { display: block }`. Explicit `w-*`/`size-*` utilities
still win.
- FA-only props migrated: `spin` → `animate-spin`, FA size strings → em
values.
- Dropped `@fortawesome/*` + `@awesome.me` kit deps, both `.npmrc`
files, and
the FA rule in `shared-vite-config.ts` (now externalizes
`lucide-react`).
- Select multiple items in the library and add tags to all of them at
once
- Tag suggestions pop up as you type, based on your existing tags
- Fixed the selection bar looking squished (buttons no longer wrap onto
two lines)
- Video generation now defaults to Seedance 2.5 (was 2.0)
- Omni Reference is the default video input mode for everyone
- Model pickers on the create pages now group models by family
(Seedance, Veo, Kling, etc.) with a flyout submenu per family, so the
list is no longer one long scroll
- The flyout opens on hover or click, and automatically scrolls the
currently selected model into view
- New setting in **Settings → General** ("Group models by family") to
switch back to the classic flat list. Grouped is the default and works
in both the webapp and the desktop app
- Model rows now use full-color brand icons, while the picker button
keeps its existing look
- Video models show capability tags: **Audio Support**, **Start/End**,
and **Reference**
- Short model lists (under 8 models) stay flat, and mobile keeps the
existing bottom-sheet list
- Fix: videos in the media viewer can now go fullscreen (the expand
button was disabled)
## Summary
New marketing page at /minimax-h3 where visitors can try MiniMax H3 for
free,
right on the page.
- Hero is a working promptbox, locked to MiniMax H3 with a "Free" badge
- Full settings available: aspect ratio, resolution, duration, sound,
start/end frames
- New users sign up in a modal when they hit Generate (Google or email),
then their prompt runs immediately without leaving the page
- Generated videos show up right below the promptbox, with a queue
notice and a CTA to keep creating in the app while they wait
- Example videos from MiniMax's announcement, self-hosted with
thumbnails, only load when played, credited to MiniMax
- Overview, highlights, prompt tips, and FAQ copy making the free offer
clear Website promptbox updated to match the current webapp promptbox
(reference deck, drag & drop, focus mode, same buttons)
- SEO, redirects, and URL aliases (/minimax, /hailuo) wired up
- Characters modal: new eye button on character cards to view the image
full size
- Mention chip menu: tapping the avatar thumbnail also opens the
full-size view
- Card buttons (view/edit/delete) are always visible on touch devices
instead of hover-only
- Mobile: Characters open in a bottom drawer with a 2-column grid,
matching the other settings drawers
- All mobile drawers now have a close (X) button
- All mobile drawers can be pulled down to dismiss - the sheet follows
your thumb and closes from where you let go
## Fixes & improvements
### Model picker
- Removed the "Seedance 2.5 - SOON" placeholder from the video model
picker; it was confusing now that Seedance 2.5 Preview is live.
### "Not all media files could be found" error
- Fixed generations failing when the same media file ended up twice in
one reference list (e.g. added as two references, or the same file
uploaded twice).
- Requests now automatically drop duplicate media references before
sending - applies to image, video, audio, mesh, and splat generation on
both the web app and the desktop app.
### Gallery picker UX
- Media already added to a slot is now greyed out with an "Added" badge
in the library picker, so the same file can't be added twice to the same
field.
- Applies per slot across the create pages (video, image, audio, 3D
object, world) on web and desktop - references, start frame, end frame,
video refs, audio refs, and multi-view slots.
- Reusing the same image across different slots (e.g. start *and* end
frame, or two 3D views) is still allowed.
### What
- Drag files onto the promptbox on the Image, Video, and Audio create
pages to add them as references - same look and behavior as the webapp
- On those pages the promptbox is now the only drop target; dropping
elsewhere does nothing
- Every other page keeps the existing upload-modal behavior
### Notes
- The webapp's drop code doesn't work as-is on desktop - Tauri
intercepts file drags before the page sees them, so desktop needed its
own path
- Both now share one component, so the two apps stay in sync going
forward
- Model limits still apply: each box only accepts the reference types
its model supports, and existing slot/duration caps are unchanged
- **Fixed: picking an audio reference on mobile opened the photo/video
picker.** iPhones don't have an audio library, so the browser fell back
to Photos - the only thing you could "attach as audio" was a video, and
it silently uploaded as an audio ref. Audio pickers now open the Files
app filtered to real audio files (MP3, WAV, M4A, FLAC…).
- **Non-audio files are now rejected with a clear error toast** instead
of being silently attached (also covers drag & drop and Android's "any
file" picker).
- **Fixed: limit warnings never appeared in the web app.** Errors like
"Total audio duration cannot exceed 15s" from shared components were
firing into a toast system the web app doesn't render. They're now
routed to the app's own toasts, so length/limit errors actually show up.
- Applies to the create video, create audio, and desktop reference deck
flows
### New stuff
- Character `@mentions` in the video promptbox now show as **chips with
a thumbnail**, so you can see exactly which character is attached to the
prompt
- Clicking a chip opens a menu to **Replace** (pick another character),
**Preview** (view their image), or **Remove** the mention
- Chips behave like a single unit - one backspace deletes the whole
mention, and removing/replacing can be undone with Ctrl+Z
- Copying prompt text still gives clean plain text (`@name ...`), and
pasting it back restores the chips
- When two characters share a name, the one you actually picked is now
the one that gets sent
### Fixes
- Mobile: the `@` autocomplete no longer goes off-screen - it stays
inside the viewport, flips below the caret when needed, and taps select
reliably
- Focus mode: the autocomplete no longer pushes the prompt sideways /
causes overflow
- Enter-to-generate now works while the mention editor is active on the
webapp
Tanish Manku
Brandon Thomas
Tanish Manku
BFlatandTanish Manku
hanashi