mirror of
https://github.com/heygen-com/hyperframes.git
synced 2026-09-04 07:19:52 +00:00
WaterrrForever
b9be0b2625
* feat(studio,cli): per-frame board comments, self-refreshing storyboard, status-aware preview landing Per-frame comment boxes on the storyboard board batch into .hyperframes/frame-comments.json (a resubmit wins per frame; unconsumed comments on other frames are kept). Submitted-but-unconsumed comments stay visible — a toolbar banner plus a per-tile echo — until the agent consumes the file; the banner also says what to do next (reply anything in the agent chat). The board keeps itself current: GET /projects/:id/signature exposes the watcher-cached project signature, the storyboard payload carries the signature it was derived from, and the view polls at 2s (hidden tabs skipped, re-checked on visibility), refetching in place with no loading flash. Posters bake the signature into their URL so tiles fill in as sketches land and a poster that failed mid-write retries on the next version; the empty state upgrades itself when STORYBOARD.md appears, and its handoff prompt now points the agent at the review loop and uses the parser's real status vocabulary (outline, not planned). preview lands the browser on the storyboard view while the board is the review surface — any frame built, or pure planning (srcs declared, none on disk yet) — and on the timeline once the video is assembled. * feat(skills): the review loop — plan, sketch, build as one shared process hyperframes-core/references/review-loop.md is the single source for the three-pass collaborative review: the plan proposed on a live board (§ 1), wireframe sketches marked built with one layout question (§ 2 — real words on plain blocks, run no CLI; a confirmed board is itself a valid deliverable when the user asked for a storyboard, not a video), the build dressing confirmed layouts (§ 3, worker or inline), and the final look (§ 4). Autonomous runs skip every gate and keep one question before render. The three narrative workflows' Steps 3/4/6 collapse to references plus their sketch stand-ins (captured-asset blocks for product-launch-video, plain code panels for pr-to-video); the confirmed-sketch handoff stays in each frame-worker prompt. general-video plans on a board for multi-scene narrative pieces in collaborative mode — its sketch pass is layout-before-animation with the user watching. The router treats "I want a storyboard" as a process request rather than a route, and closes exploratory intake by recommending a route plus how the run will review. The supporting contracts land next door: the comments channel (silent submit, one reply picks it up, check the file before the words) in brief-contract § 1; the sidecar schema and the built status rung in storyboard-format; the mode question asked first and alone in the three workflows' Step 0. * feat(media-use): user memory — remembered preferences and frozen recipes Two tiers of memory on media-use's existing two-tier storage split. Preferences (lightweight): confirmed brief answers — destination, aspect, language, mode, voice, style preset — recorded to the project's .media/preferences.json (committed, the team inherits it) and promoted to the personal ~/.media/preferences.json once the same value is confirmed in two different projects (a sightings ledger accumulates the cross-project evidence user-side, since project files can't see each other). prefs.mjs get/record; merge reads project-over-user; a changed value restarts its provenance. Recipes (heavyweight): one approved run frozen as a named, versioned bundle — frame.md, the storyboard skeleton (structure kept: durations, transitions, srcs, Video direction; statuses reset to outline; content blanked to per-frame fill-ins naming the beat's role), and the confirmed brief values. Named folders, not content hashes: re-freezing bumps version and archives <name>@v<N>; a freeze is already confirmed, so it promotes to the user tier immediately. recipe.mjs freeze/list/use, plus resolve --type recipe --entity <name> delegating like grade/lut. 16 new node --test cases; the media-use lib suite is 168/168. * feat(skills): wire user memory into the brief and the review loop brief-contract § 2 gains Remembered defaults: read the merged preferences before Round 2 and let a remembered value become the recommended option with a receipt naming its source project. Memory changes the default, never the question — every ask-marked field still gets asked, and what the request says this time beats what was picked last time. Record only what the user actually confirmed (a defaulted voice nobody chose is not an answer; a "go" that accepts the recommended defaults is). The first record announces itself once; after that the receipts carry the reminder. In autonomous mode a remembered value becomes the decided value, receipt included. The three narrative workflows read the remembered defaults before Round 2, record the confirmed answers at the Step 0 gate, record the chosen preset at the Step 2 gate (pr-to-video excepted — its preset is fixed), and fall back to the remembered voice when the request names none. general-video's discovery reads the same defaults. Recipes wire in at both ends: Step 0 checks for a matching recipe before the mode question — one question, plural-aware, and adopting one fills the brief, skips the design step, and drafts the storyboard from the frozen skeleton while every review gate still runs. The review loop's final look (§ 4) offers the freeze once after approval, and the confirmation teaches the recall phrase — the name is something the system reminds the user of, never something they must remember. The router recognizes a named recipe or "like last time" as a route. * docs(skills): the sketch pass names check, not the deprecated validate * feat(skills): intent-layer references — process, route briefs, capability menu, BRIEF.md format * feat(media-use): brief skeleton as the recipe's fourth artifact; flow/storyboard preference keys * feat(skills): the intent layer conducts every brief — workflows execute BRIEF.md * feat(skills): retire the mode preference key; sync catalog surfaces for intent layer * refactor(skills): dedupe router vs intent-layer guidance — one owner per rule * feat(skills): the design ask — own spec, pick by eye from showcases, or defer * docs(skills): the design ask says the honest line on capture routes * feat(skills): product-launch-video absorbs website-to-video as the tour angle * refactor(skills): keep product-launch-video pristine — a tour is brief intent, not a pipeline branch * feat(skills): production loop + genre lenses; general-video goes freeform (route yours, laws hold) * refactor(skills): /hyperframes is the front door - route tables and scope lists leave the workflows * docs(skills): review-loop pass across skill catalog * fix(cli): pass project dir to openStudioBrowser in background-server path * feat(skills): add pitch-round reference - verbalized sampling concept gate * feat(skills): wire pitch round into intent layer - completeness triage + route eligibility * feat(skills): editorial capability recommendations, handoff disciplines, menu-probe split * feat(skills): pitches carry their machinery; source-only-formed requests pitch the telling * feat(skills): companion goes director - ceiling treatment plus blueprint/rule citation discipline * fix(scripts): sandbox npx-leak guard - private npm global prefix keeps npx on the branch CLI * chore(skills): resync manifest hash after formatter pass reflowed general-video tables * fix(skills): recipe freeze reads workflow from BRIEF.md; style_preset records require workflow scope Two holes found by a live companion-run freeze: the agent-supplied --workflow contradicted the run's actual workflow (recipe.json said faceless-explainer, brief-skeleton said general-video), and the style_preset lookup missed because the preference had been recorded under the bare key. - freezeRecipe resolves the workflow from BRIEF.md frontmatter; the flag is a fallback for briefless projects and a contradicting flag is ignored (noted). - recordPreference refuses a bare style_preset — the scoped key is the only writable shape; freeze tolerates legacy bare records via read fallback. - review-loop § 4 / media-use SKILL / brief-format wording follow the machinery.
51 lines
10 KiB
Markdown
51 lines
10 KiB
Markdown
# device-surface-showcase — Device / Surface Showcase
|
||
|
||
**intent**: A product surface — a device mockup or a floating browser/app window — is the hero held in frame while its screens cycle through a real flow, showcased by a camera move that ranges from a static hold to a continuous 3D push.
|
||
|
||
**roles served**
|
||
|
||
- Key*Feature (from key-feature-device-screen-tour, key-feature-floating-window-scroll, key-feature-3d-device-hand-demo): show a feature being \_experienced inside its real interface* — the surface houses the action and its screens advance through a flow, rather than enumerating tiles or chasing a cursor across a workflow. (Note: all three drafts are Key_Feature; this blueprint is role-narrow but mechanic-rich — variants differ by MECHANIC, not role.)
|
||
- Key_Feature (from demo-page-scroll-spotlight): the floating-window push-scroll variant carried to a spotlight climax — a real webpage rendered as a tilted 3D card coasts in (power2, like a phone held up — no spring), header keywords flare on a karaoke glow as the VO names them, the page rolls to the demoed section, and one element LIFTS off the surface (translateZ + scale) under a radial spotlight that dims the rest.
|
||
|
||
**duration**: 5–9.6s (page-scroll-spotlight 5–9s · floating-window 7.8s · 3d-hand 7.9s · device-tour 9.6s)
|
||
|
||
**shot structure** One product surface — a `[device mockup]` or a `[floating browser/app window]` — is the persistent hero on a `[styled backdrop: gradient / radial / stylized 3D void]`; its `[screens/sections]` cycle through a real `[product flow]` while a showcase camera (static-hold, push-in→zoom-out, or one continuous push) presents it. Each screen state holds ~1.0–1.5s.
|
||
|
||
- Scene 1 (0.0–~1.5s): The surface ESTABLISHES — it `[slides in from an edge / drifts in from a tilt / dissolves from a full-frame title card]` and settles, with a `[accent shape or backdrop]` resolving behind it; the first `[screen]` is visible. The showcase camera begins (see variants).
|
||
- Scene 2 (~1.5–~Xs): The surface is OPERATED on its own face — a `[tap/select/scroll]` triggers the first screen advance: old content `[pushes out / scrolls up]`, new `[screen/section]` `[pulls up / pushes in from the side]`; concurrently a `[label / header word / side headline]` updates. The camera continues its move.
|
||
- Scene 3+ (~Xs–end, repeat for `[2–4 screen beats]`): The surface ADVANCES through successive `[screens/sections]`, each a discrete swap or scroll synced to the surface's flow, while the secondary copy `[swaps out-up / in-up]` or stays marked to hold reading position. HOLDS on the final `[screen]` (or, for one variant, blooms out — see variant).
|
||
|
||
- Variant — static-tour (key-feature-device-screen-tour, 9.6s): a `[device mockup]` slides in from off-screen and settles (ease-out); an `[accent-color shape]` scales up behind it (spring overshoot). Camera STAYS STATIC the entire clip — all motion is element/UI-level: a tap COMPRESSES a button (95%→100%), the UI scrolls/transitions to the next view (old pushes out, new pulls up), and a `[side headline]` SWAPS beside the device (old slides up + fades, new slides up + in) per screen. Holds on the final screen. No camera move, no cursor.
|
||
- Variant — floating-window (key-feature-floating-window-scroll, 7.8s): OPENS on a full-frame `[title card]` (a small `[icon]` draws in at center, `[feature name]` below; holds ~2s), which DISSOLVES to a `[macOS-style browser/app window]` floating on a `[vivid gradient]` (traffic-lights + `[URL pill]` + tabs; left nav, central content, right `[sidebar]`). Camera PUSHES IN on a `[target region/sidebar]` (active item highlighted `[accent]`, a cursor drifts down the list), then ZOOMS BACK OUT to re-frame the whole window while the content SCROLLS through `[sections]`; the `[highlighted item]` stays marked. One push-in→zoom-out arc, gated by the title-card opener.
|
||
- Variant — 3d-hand (key-feature-3d-device-hand-demo, 7.9s): FULLY 3D — a `[3D device]` drifts in a `[stylized 3D void / bloom + particles]`, opening tilted and self-rotating to face the lens nearly flat as ONE CONTINUOUS forward camera push begins (no cuts). A glossy `[3D hand]` rises from the bottom-foreground and GESTURE-DRIVES the surface: it swipes to scroll a `[picker/sidebar panel]` of `[option cards]` and taps `[option]` (while a `[header word]` letter-flips in place); the selection APPLIES — a `[new layout]` grows from center to fill the device face, nav flips, a `[marquee]` scrolls horizontally; the hand swipes again to scroll the page upward through `[sections]`, then drifts out. The camera never stops pushing; the bright device face keeps growing toward the lens until it BLOOMS into a `[light]` wash — a zoom-through "portal" exit that fills the frame.
|
||
|
||
**motion vocabulary** surface establish (edge slide-in + settle / tilt drift-in + self-rotate-to-camera / title-card dissolve); accent shape spring behind surface; element-level screen-cycling (scroll-swap, push-in-from-side, scale-swap); button tap-compress; staggered side-headline reveal + copy swap (out-up / in-up); in-place header-word letter-flip; floating browser-window-on-gradient idle float; full-frame title-card opener (icon draw-in + label); camera push-IN on a region; camera zoom-OUT re-frame; content scroll-through; one continuous 3D camera-follow push (no cuts); 3D device drift + self-rotate; stylized-environment bloom/particles; 3D-hand entrance + swipe-scroll + tap (gesture-driven); picker-panel slide-in; template-apply grow-from-center; horizontal marquee scroll; gesture-driven page scroll; zoom-through bloom/portal exit; static-hold (no camera) as the floor of the camera range.
|
||
|
||
**rule mapping** (per motion verb → backing rule, or flagged special)
|
||
|
||
- screen-cycling — UI scrolls/sections scroll inside the surface (device-tour, floating-window scroll, 3d-hand page scroll) → `3d-page-scroll` (webpage/app as a tilted card whose content `translateY`-scrolls to sections; primary mechanic for the surface's screen flow)
|
||
- floating-window establish + the surface presented as a tilted/floating UI card → `3d-page-scroll` (the tilt/perspective framing) + `css-3d-transforms` (perspective/`translateZ` depth)
|
||
- screen / side-copy state swaps (discrete screen states; side headline content swapping per beat) → `discrete-text-sequence`
|
||
- side-headline reveal (staggered fade + slide-up) → `discrete-text-sequence`
|
||
- in-place header-word letter-flip (3d-hand) → `hacker-flip-3d`
|
||
- screen swap as a coordinated shrink-out / pop-in between two screen states → `scale-swap-transition`
|
||
- template-apply "new layout grows from center to fill the face" (3d-hand) → `center-outward-expansion` (clustered-at-center → expand to fill)
|
||
- the surface morphing between states / title-card→window dissolve as the eye-anchor transition → `card-morph-anchor`
|
||
- button tap-compress (95%→100% press feedback) → `press-release-spring` (or `physics-press-reaction` for a heavier press)
|
||
- floating-window cursor click on the highlighted list item → `cursor-click-ripple`
|
||
- accent-highlight pop on the active sidebar/list item → `asr-keyword-glow` (accent glow on the focused item)
|
||
- drifting cursor down the sidebar list (floating-window) → `camera-cursor-tracking` (flat-cursor drift; pairs with the push-in)
|
||
- floating browser-window idle float / 3D device drift-breathe → `sine-wave-loop`
|
||
- 3D device drift + self-rotate-to-camera + perspective depth (3d-hand) → `css-3d-transforms` (CSS-3D) **or** `3d.md` technique (true Three.js/R3F device); see camera modifier
|
||
- horizontal `[marquee]` scroll (3d-hand) → `viewport-change` (PAN mode on the marquee strip) — _thin fit; a literal CSS-marquee/translateX loop is closer to a `gsap-effects`/CSS recipe than a named motion rule_
|
||
- 3D-hand entrance + swipe + tap as the interaction DRIVER (gesture input that scrolls/selects) → **flagged special — needs a heavier capability beyond the rule library (R3F/Three.js + WebGL), NOT a motion-shape rule.** The 3D hand model + WebGL bloom have a _technique_ backing (`3d.md` — R3F, `useGLTF` HandModel, `--gl=swiftshader` for the shader/bloom), but no motion-shape rule models a 3D hand as the swipe-to-scroll / tap-to-select gesture protocol. `context-sensitive-cursor` / `camera-cursor-tracking` only model a flat typing/pointer cursor, not a 3D gesturing hand.
|
||
- zoom-through bloom / portal exit (3d-hand) → **flagged special — needs a heavier capability beyond the rule library (WebGL), NOT a named transition rule.** Capability is `techniques.md` → WebGL shader (via `3d.md` headless WebGL: `--gl=swiftshader --concurrency=1`), but no named transition rule covers a bloom/portal fly-through.
|
||
|
||
**camera modifier**: The showcase camera spans a RANGE keyed by variant, all on a single content-wrapping virtual camera (`viewport-change`):
|
||
|
||
- static-tour → NO camera move (`viewport-change` held at scale 1, or omitted); all motion is element-level. This is the floor of the range and what distinguishes the device-tour from the rest.
|
||
- floating-window → a two-phase push-in → zoom-out arc → `multi-phase-camera` (e.g. dramatic-reveal 1.1→1.0→0.95 feel): push IN on the `[sidebar/region]` via `coordinate-target-zoom` (off-center target = scale + counter-translate), then `multi-phase-camera` zooms back OUT to re-frame the whole window while content scrolls.
|
||
- 3d-hand → ONE continuous forward push (no cuts) → `multi-phase-camera` in steady-push mode (1.0→1.03→1.06… plus its sine micro-drift) layered over `css-3d-transforms`/`3d.md` so the device self-rotates-to-lens during the push; the push runs unbroken into the bloom/portal exit (exit itself is the WebGL-shader flagged special above). Across all three: `viewport-change` is the base virtual-camera primitive; `multi-phase-camera` sequences the push/zoom phases (and supplies the always-on micro-drift that keeps even the "static" tour from feeling dead); `coordinate-target-zoom` aims the push at off-center screen detail.
|
||
|
||
**Overflow (pan/scroll surfaces — required for a clean `check`):** a panned or scrolled surface deliberately moves content PAST the edges of its framing card. Clip it at the card (`overflow: hidden` on the card/window) AND mark the moving inner layer (the `.world` / surface wrapper holding the screenshot + any markers/labels) with `data-layout-allow-overflow` — otherwise `check` reports `text_box_overflow` / `container_overflow` errors for the parts that scroll off (e.g. a marker label panned off the left edge). The card clips them visually; the attribute tells the layout audit it's intentional, not a layout bug.
|