Repository navigation
feat(website): Cinematic Studio opens on a playground-style side panel - #19316
Conversation
The side panel becomes the default layout: Model, Scene, one Shot card listing the camera and every direction part, References and Format. The bottom composer uses segmented controls whose parts open their own pickers or menus, a References menu behind the +, and shared tooltips. Camera options are drawn as shaded objects and direction options as neutral line icons. Credit amounts stay hidden on both layouts for now. Same tree as #19227, rebuilt as one commit on main.
|
The latest updates on your projects. Learn more about Vercel for GitHub. 2 Skipped Deployments
|
|
Navigate logical layers of code changes, visualize relationships, and explore their blast radius. No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository: Comfy-Org/ComfyUI_frontend/.coderabbit.yaml Review profile: ASSERTIVE Plan: Advanced Run ID: 📒 Files selected for processing (40)
💤 Files with no reviewable changes (2)
Included review availability: This review used your included allowance. Your plan provides up to 8 included reviews per hour; 3 remain after this review. 📝 WalkthroughWalkthroughCinematic Studio updates its shot, format, reference, and enhancement controls. It adds camera and direction illustrations, changes picker behavior, and adjusts studio layouts and credit display. Tests cover the updated controls and interactions. ChangesCinematic Studio refresh
Priority: ➖ Normal Estimated code review effort: 4 (Complex) | ~60 minutes Change: Feature Sequence Diagram(s)sequenceDiagram
actor User
participant CinematicShotList
participant CinematicStudioPanel
participant CinematicPicker
User->>CinematicShotList: Open a shot-setting row
CinematicShotList->>CinematicStudioPanel: Emit the picker key
CinematicStudioPanel->>CinematicPicker: Open and position the picker
User->>CinematicPicker: Select an option
CinematicPicker->>CinematicStudioPanel: Pass the selection to choose
Possibly related PRs
Merge Risk: ⚪ Minimal · up to The updated controls keep take selection within the supported range and hide credit amounts in both layouts. No identified issue needs to be resolved before merging. Security Architecture ReviewSecurity architecture risk: 🔵 Low · up to The new layout and hidden credit amounts do not appear to change who can submit a generation request or how an in-progress request is handled. Some downstream enforcement was outside the reviewed change, so the assessment retains limited uncertainty. Retained concerns Security review detailsSecurity Blast Radius
Trust Boundaries and Controls
Resilience and Maintainability Implications
Important Pre-merge checks failedPlease resolve all errors before merging. Addressing warnings is optional. ❌ Failed checks (1 inconclusive)
✅ Passed checks (6 passed)
Full details: Website End-To-End Regression CoverageExplanation The PR changes observable website runtime behavior in Resolution Provide the full PR description or confirm that its omitted section contains no concrete explanation for not adding a Playwright regression test. If no explanation exists, add or update a relevant Playwright test under ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
benjcooley
left a comment
There was a problem hiding this comment.
Verified replacement for #19227. The clean PR has the same 40-file change set and stable patch ID as the fully reviewed predecessor, now as one Mariana-authored commit on main with no AI co-author trailer. No design or code changes were introduced by the replacement.
benjcooley
left a comment
There was a problem hiding this comment.
👨💻 BILL — exact-head review of #19316 at b5cfc841ba: READY.
| Severity | Count |
|---|---|
| BLOCKER | 0 |
| SHOULD FIX | 0 |
| TRIVIAL | 0 |
Same change as #19227, with clean history
- Authorship: one commit, authored and committed by Mariana Martinho, with no
Co-authored-byor AI trailer. - Identical change: the stable patch-id is
a0bd7cf433c42a318431b19703cd312d27b7f948, identical to #19227's final headcebc807ab1, each measured from its own merge base. Both change the same 40 files (+2269/−967). - Merges cleanly with current main (
dab0dbee1c).
One commit was new to me. My last #19227 verdict (0/0/0) was at b60b0ebf3f. The final #19227 head adds one more commit, cebc807ab1 "draw direction options as neutral line icons", so #19316 contains it. I've reviewed it now for correctness and accessibility; the drawings are Mar's:
CinematicDirectionIcon: the SVG isaria-hidden, and its gradient, pattern and filter ids come fromuseId(), so they're unique per instance and SSR-stable. The segment buttons keep theiraria-label(title: label).- No empty icons: the segments bar and the Shot list render icons only for
[...lookGroups, gradeGroup].direction-icons.test.tsasserts every look option has a unique drawing. Every grade option carries a palette, so grade always takes the swatch branch. Camera parts keep their own previews and never go through this path. - Option grid: the corner icon shows only when an option has a preview and a drawing, and never for
auto.
Verification: all 19 Cinematic Studio test files (171 tests) pass locally on this head. CI when I posted: 5 pass, 27 pending. Merge on green.
🌾 Fallow: ✅ PassedFallow audits the diff for duplication, complexity and dead code. It gates on new findings only, so when the audit completes, a red is something this PR introduced or moved — never inherited backlog. A red can also mean the audit itself did not finish; the details below say which happened. No new findings in the changed files. Reproduce locally# the base the gate compared against
npx fallow audit --root . --quiet --format json \
--changed-since 8d3a084a07a62dfb7d84ae747db6f1e25d4349b5
Full docs: ADR-DEVEX-LINT-0015 🌐 Website E2ETip All tests passed.
🧩 Custom Nodes Ecosystem Matrix: 🟡 Withheld — could not measureExecutes the frontend JS of every registry pack that ships any, against this change, in both renderers. The eight This candidate changes nothing the matrix measures, so every expensive stage was skipped. Reproduce locally# build one shard's fixture, then run it
python3 scripts/registry-census/build_matrix.py --shard 1/4
MATRIX_RENDERER=vue MATRIX_VUE=1 \
npx vitest run --config vitest.matrix.config.mts
Full docs: scripts/registry-census/README.md 🔗 Website PreviewWebsite Preview: https://comfy-website-preview-pr-19316.vercel.app Models is built and gated by PostHog (workshop-enabled), using staging Cloud. This commit: https://website-frontend-he7xsy4ds-comfyui.vercel.app Last updated: 2026-09-28T23:40:18Z for |
… models, animate a still (#19370) ## Summary Stacked on #19326 (base branch `rob/cinematic-studio-luis-port`); it retargets to `main` once #19326 merges. Both are rebased onto Mariana's side-panel redesign (#19316): her layout, menus and clay camera icons are the base, and video fits into her components. It adds Luis's Image/Video mode to Cinematic Studio on main's Router path. The video models go through the same `router_render`, credit gate, parallel takes and retries as image shots, and their requests are built with `studioRouterForm`. - **Image / Video switch** above the side panel (and above the composer in the composer layout). Each mode keeps its own model. - **Ten video models, in picker order:** Seedance 2.5, Seedance 2.5 Edit, Seedance 2.0, Seedance 2.0 Fast, Seedance 2.0 Mini, Wan 3.0, FLUX.3 Video, Gemini Omni Flash 1.1, Grok Imagine 1.5, Kling 3.0. Each has picker tags for its sharpest resolution, length range and sound. - Each text-to-video model is paired with its image-to-video operation. Adding a starting frame moves the shot there, the same way references move an image shot to its edit operation. - Kling 3.0 is text-only until its image-to-video is runnable on the Router. - Veo 3 is left out on purpose. - **What a shot can choose comes from each contract.** Lengths, resolutions, frames, sound, starting/ending frame and source video are all read through the same shared Router mapping that sends the request, so the picker only offers what reaches the model. That also covers models whose fields are named differently (Veo, Wan, Gemini) where Luis's parser missed them. - **Video controls in Mariana's components:** - Output (side panel) and the composer's Format segments show frame, resolution, length and "Generate audio" (off by default), sharing one `useVideoMenus` helper the way her image menus share `useFormatMenus`. - References (side panel) and the composer's "+" menu hold the starting and ending frame, or the video to edit (Seedance 2.5 Edit), next to Colors. One `referenceSlots` helper decides which slots a model gets in both layouts. - Her AI prompt switch shows a motion hint in video mode. - **Playback:** video takes play in the frame and show as a muted thumbnail. "Use as reference" is offered on stills in image mode only. "Animate image" starts a video from a still. - **One clip per video shot, and no price shown.** The pricing rules have nothing that matches a studio clip (only Grok image-to-video, at a fixed 6 s / 480p), so the credit gate only asks for a positive balance. ## Studio polish - **A finished take keeps its frame and column width until its image or video has loaded.** Before, the stage collapsed or shrank to a small box, and the filmstrip jumped. - **Filmstrip thumbnails share one 3:2 shape**, cropped to fit, whatever the take's frame. - **The progress label reads "Generating image" or "Generating video"** instead of "Rendering take A". - **On a video take, the take details and actions sit at the top of the frame**, so they don't cover the browser's play bar. ## Reusing a still ("Animate image", "Use as reference") Both buttons used to download the take in the browser. Most providers leave the image on their own storage without a CORS header, so the download failed and the buttons silently did nothing. We checked this from a localhost page with `fetch` and with header checks: | Where the still lives | Models | Page can read it? | |---|---|---| | Inside the reply | GPT Image 2 / 2.5, Nano Banana Pro / 2 | yes | | Comfy storage (Router re-host) | FLUX 2, Grok | comfy.org only (not localhost or previews) | | BytePlus | Seedream 5.0 Pro | no | | DashScope Singapore / Beijing | Qwen | no / yes | | Recraft | Recraft V4.1 | yes | A reused take is now kept as a reference to the output: its link, plus the picture when the page could read it. - **Link:** sent to models whose contract takes URLs, since their server fetches it and nothing is uploaded from the page. That covers Seedance 2.5 / 2.0 / Fast, Wan, Gemini and Grok image-to-video, and Qwen references. - **Picture:** sent to models that need the bytes. When the page couldn't read the picture, Generate is held back with a note instead of failing silently. **Known limitation: animating a GPT Image or Nano Banana still.** These models return the picture inside the reply, so there's no link to pass on. - **Seedance / Wan / Gemini / Grok** take the starting frame as a link, so the still is uploaded to Comfy storage first. Comfy storage only allows uploads from comfy.org, so this fails on localhost and previews ("Your file could not be uploaded"). It works on comfy.org. - **FLUX.3 Video** takes the picture inside the request. A large still (a 3840×2160 GPT PNG) can go over the site's 10 MB request limit. Seedream's Router contract doesn't expose `response_format`, so asking it for `b64_json` isn't possible from the studio. ### Long term (not a blocker, discussion in Slack) The durable fix is for Comfy to keep every Router output and let the Router take it back as an input: 1. Store every output in Comfy storage or saved assets (`comfy_save_asset`), as FLUX / Grok / Seedance video already are. 2. Accept an asset id or Comfy-signed URL for `reference_images` / `first_frame`, resolved server-side. Then no image passes through the browser at all. `TakeImage` is shaped to become that asset id. Smaller asks: localhost and preview origins on `api-nodes-prod` CORS, and `response_format` on the BytePlus image contracts. ## Testing on localhost - **Fully testable:** all text-to-video; FLUX.3 and Seedance 2.0 Mini image-to-video (the frame is sent inline); and animating a still into the link-taking models. - **Needs comfy.org:** uploading your own file as a frame for Seedance / Wan / Gemini / Grok, and the Seedance Edit source video. These go through Comfy storage, whose CORS is comfy.org-only. ## Test plan - [x] `video.test.ts`: capabilities read from real contracts (lengths, resolutions, frames, audio, first/last frame, source video, frame choice) and request parameters for each kind. - [x] `take-image.test.ts`: keeping a take (link, picture or both), what each model receives, and which contracts accept links. - [x] `CinematicStudio.test.ts`: a clip in video mode, animating a still, animating a provider-hosted still by link, the "needs the picture" note, and reference hidden in video mode. `CinematicTakeMedia.test.ts`: video playback. - [x] Offline request preparation for all 18 video operations (network blocked): each prepares a valid request with the watermark off. - [x] `CinematicTakeFrame.test.ts`: the frame and column hold until media loads, the Generating image/video label, even thumbnails, and the take details at the top of a clip. - [x] `models.test.ts`: `videoShotBlock` (source video needed, no starting frame, unreadable frame). - [x] Rebased on #19326 + current main. Website suite: 514 files, 9,775 tests passing. Typecheck, knip and the fallow audit pass. - [x] Browser, side panel (default): mode switch; the ten video models in order with tags; video Output shows 16:9 · 720p · 5s · Generate audio; References shows Starting frame + Colors; Seedance 2.5 Edit shows "Video to edit" and holds Generate with "Add the video to edit first."; image mode unchanged. - [x] Browser, composer layout (`?ux=e`): video Format segments; the "+" menu lists Starting frame + Colors in video mode and Character, Palette, Colors in image mode. - [x] Earlier, before the rebase: Generating image/video label, 84×56 thumbnails, Animate / Use as reference with demo stills, take details at the top of a clip. - [ ] Live signed-in video runs (spend credits). 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: luisvitela-hub <luisvitela-hub@users.noreply.github.com> Co-authored-by: benjcooley <benjcooley@gmail.com>
…, a Colors palette, and request fixes (Comfy-Org#19326) ## Summary Ports self-contained pieces of Luis's `cinema-studio-integration` work onto main, fitted into Mariana's side-panel redesign (Comfy-Org#19316). Her layout, Format menus, References menu and clay camera icons stay as they are, and main's Router path and credit estimates are untouched. Six commits: image models, dropping Seedream 4.5, the Colors palette, two request fixes (no watermark; no catalogue examples in studio requests), and tests for sampling colours. ### 1. Image models and frames - Cinematic Studio now offers **13 image models** (was 5). It adds Seedream 5.0 Pro, FLUX 2 Max, Nano Banana 2, Qwen Image 3.0, GPT Image 2, GPT Image 2.5 Flare and Sunburst, Grok Imagine Image 2.0 and Recraft V4.1, and **drops Seedream 4.5**. Seedream 5.0 Pro is the new default. It has no 21:9 size, so the studio opens on 16:9, and starter shots asking for 21:9 move to it. Video stays out of the studio for now. - Adds **3:2 and 2:3** frames. - **Each model only offers the frames it can make.** The frame menus in Mariana's Format controls (side panel and composer segments) share one filter in `useFormatMenus`. - **Frames per model:** The frames are read from the model's Router contract (its size menu). Switching to a model that can't make the chosen frame moves it to the nearest one. With references attached, the frames follow the edit operation that will actually run. - **The studio asks for the exact size** of that frame (`model_specific`), and the credit estimate prices that same size. - **Reference support is read from each contract**: which operation keeps references and how many images it takes. The existing "{model} can't use references" block now also covers attaching more images than a model accepts. - **No Router or Models-tab changes.** Everything goes through the existing `router_render` path, credentials and credit gate. The contract-reading logic is ported from Luis's `cinema-studio-integration` branch. ### 2. Colors palette (sent as words) - A **Colors** control holds up to eight swatches: add, change, remove, or mark one as the main color. In the side panel it sits in References, below Character and Palette. In the bottom composer it's a **Colors** item in the References (+) menu that opens a small Colors panel. - **From an image** fills the swatches with a picture's dominant colors. The picture is read in the browser and never uploaded. It's resized without smoothing, so only colors actually in the picture count. - The colors reach the prompt as words: `Color grade: dominant colors #204060, #e06060, #ffa060. Keep skin tones believable.`, plus a `Palette priority: … is the main color` line when one is marked. So they add no reference image, cost nothing extra and work with every model, including those that can't take references. - The existing Palette image slot is unchanged. Both can be used together. - The prompt wording and color counting are ported from Luis's branch. ### 3. Two request fixes - **No watermark.** Seedream 5.0 Pro, both generate and edit, stamps a visible watermark unless asked not to, because its Router contract defaults `watermark` to true. Studio requests now turn off every watermark switch a model's contract has, found by name. - **No catalogue examples in studio requests.** `router_render` starts a request from the model page's first example unless it's given a `form`. The studio didn't intend that, but it happened: - FLUX 2 Max attached its example sofa photo to every take. - GPT Image asked for `quality: high`. - Krea asked for `creativity: low`. - The edit operations carried an example input until a reference replaced it. Studio takes and their credit estimates now pass their own form: the model's contract defaults with the shot on top. `request.test.ts` checks every studio model and edit operation for catalogue links, stray files and watermarks left on. It fails on FLUX 2 Max if the old behaviour is put back. ## Why (image models) Before this change the studio sent `{aspect_ratio, resolution}` and let the shared Router parameter mapping pick the model's nearest size **by pixel count, not by shape**. From main's own mapping: Seedream 4.5 is no longer in the studio, but it shows the problem most clearly: | Model | Chosen | Sent before | Sent now | |---|---|---|---| | Seedream 4.5 | 1:1 · 1K | 2304×1728 (4:3) | 2048×2048 | | Seedream 4.5 | 16:9 · 2K | 2496×1664 (3:2) | 2560×1440 | | Seedream 5.0 Pro | 16:9 · 2K | 2048×2048 (square) | 1920×1080 | | GPT Image 2 | 4:3 · 2K | 2048×2048 (square) | 4:3 not offered | ## Per-model frames | Model | Frames | References | |---|---|---| | Seedream 5.0 Pro | 16:9, 3:2, 2:3, 1:1, 9:16 | via edit operation | | Nano Banana Pro / 2 | all 7 | via edit operation | | FLUX 2 Pro / Max | all 7 (free width/height) | itself | | Qwen Image 3.0 Pro / 3.0 | 3:2, 2:3, 1:1 | itself | | Krea 2 Large | all 7 (21:9 sent as its 2.35:1) | none | | GPT Image 2 / 2.5 Flare / 2.5 Sunburst | 16:9, 3:2, 2:3, 1:1, 9:16 | none: the Router has no image input for them yet (see below) | | Grok Imagine Image 2.0 | all but 21:9 | none: the Router has no image input for it yet (see below) | | Recraft V4.1 | 16:9, 3:2, 2:3, 1:1, 9:16 | none | A listed size counts as a frame when it's within 3% of it. For example, Recraft's 1344×768 is 16:9 to 1.6%, and Krea's 2.35:1 is 21:9 to 0.7%. Frames that are genuinely different shapes stay hidden: 4:3 and 3:2 are 12% apart. ## Open discussion (not a blocker): how Comfy Router maps aspect ratios The shared Router mapping (`closestOption` in `config/router-parameter-options.ts`) resolves a requested `aspect_ratio`/`resolution` to the nearest listed size **by pixel count**. For a single model page that's a reasonable fallback. For an app like Cinematic Studio, which promises a frame across many models, it silently changes the shape of the picture (see the table above). The user asked for 1:1 and got 4:3, and nothing told them. Model-specific size menus also differ a lot, e.g. Recraft's 1344×768 vs a true 16:9, or Krea's 2.35:1 vs 21:9. This PR works around it inside the studio by reading each contract's size menu and sending an explicit size. It might be worth discussing whether the Router (or the shared mapping) should: 1. match by shape first and then by size, or 2. report back when it had to change the shape, or 3. expose each model's supported aspect ratios as data, so apps don't each have to parse size menus. Happy to follow whatever direction the Router owners prefer. None of this blocks the PR. ### Related: GPT Image and Grok image edits are not on the Router yet GPT Image 2, 2.5 Flare and 2.5 Sunburst can edit images and take reference images (OpenAI's edits endpoint and Comfy's own GPT Image nodes support this). But the Router operation `/v2/models/openai/gpt-image-2` (and the 2.5 variants) only accepts generation fields (prompt, size, quality, background, output format). That's why their edit pages were turned off in Comfy-Org#17951. Grok Imagine Image 2.0's edit operation is in the same position. Because the studio reads reference support from each Router contract, these models show "can't use references" when a Cast or Palette image is attached. When the Router adds image inputs for them, a contract refresh should turn references on in the studio automatically, plus a one-line `REFERENCE_OPERATIONS` entry if the edit ends up as a separate operation. ## Also worth knowing - **Reference uploads can't be tested on localhost.** Nano Banana Pro/2 and Qwen 3.0 / 3.0 Pro upload references to Comfy storage (`api-nodes-prod`) before running. That bucket's CORS only allows `https://comfy.org` and `https://platform.comfy.org`, so the browser blocks the upload from any `localhost` port. This is the same for the Models tab on main. Seedream and FLUX send references inline, so they work locally. On comfy.org (and www, which redirects there) uploads work. It might be worth adding localhost and preview origins to that bucket's CORS so reviewers can test uploads; that's a non-blocking infra question. - **Credit estimates:** 3 of the 13 models have one (Seedream 5.0 Pro, FLUX 2 Pro, FLUX 2 Max). The rest show "Cost varies", the same as Nano Banana Pro, Krea and Qwen Pro already do on main, because the estimate only covers models with verified pricing rules. The credit gate still requires a positive balance. - **Links from model pages:** the 9 new model pages now link to the studio (via `cinematicStudioHref`). The Seedream 4.5 page no longer does. ## Test plan - [x] `colors.test.ts`, `CinematicColors.test.ts` and `prompt.test.ts` cover the prompt words, the main-color line, the 8-color cap, near-shade counting, and add/remove/main/clear. They also check that sampling from an image fills the swatches and that an unreadable image shows an error. - [x] New `frames.test.ts` runs against real Router contracts. It covers the frames per model, the size sent (Seedream 1:1 → 2048×2048, Recraft 16:9 → 1344×768 through the real mapper), free-size models, reference capacity, nearest-frame fallback, and the reference-count rule. - [x] Rebased on current main (with Comfy-Org#19316). Full website suite: 510 files, 9,746 tests passing. Typecheck, knip and the fallow audit pass. - [x] Browser: the side panel (default) lists 13 models; the Format menu shows Recraft's 5 frames only; Colors appears in References (panel) and in the + menu (composer), where it opens the Colors panel. - [x] Checked in the browser, Colors: sampling a three-color picture gives exactly those three colors, the star marks the main color, and the credit estimate is unchanged. The control shows in both the popover layout and the side-panel layout. - [x] Checked in the browser, image models: - The model menu lists 13 models. - Recraft's Format menu lists only its 5 frames. - Switching from 21:9 to Recraft or GPT Image moves the frame to 16:9. - With a Cast image attached, GPT Image 2 shows "can't use references" and Generate stays blocked. - [ ] Live signed-in run on a few new models (spends credits; not done yet). 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: luisvitela-hub <luisvitela-hub@users.noreply.github.com> Co-authored-by: benjcooley <benjcooley@gmail.com>
Summary
Replaces #19227 with the same tree as one commit on
main. #19227's history still carried the pre-squash PP Formula commits from the old Hub stack (authored as an agent identity), which fails the co-author check now that it targetsmain. Nothing in the diff changed. #19227 has its review history.Cinematic Studio opens on the side panel layout (D) by default. The panel follows the compact list design: Model, Scene (AI prompt switch in the box), one Shot card listing Camera and every direction part, References, and Format as three menus above the run button. The bottom composer (
?ux=e) uses segmented controls whose parts each open their own picker or menu, a References menu behind the +, and shared tooltips. Camera options are drawn as shaded clay objects; shot, light, film and look options as neutral line icons. Credit amounts stay off both layouts for now.Changes
CinematicStudioPage.vue: default layoutd(also in production;?ux=estill opens the composer on previews); passes:show-credits="false"to both layouts.showCreditsprop (defaulttrue) threaded down toCinematicGenerateAction; when false the estimate, shortfall note and reduce button are hidden and a no-credits gate keeps only the buy button (note as tooltip fromlg; other notes stay inline).CinematicPanel.vue: Model, Scene, Shot, References, Format with small sentence-case labels; the model trigger drops the provider name (cinematic.model.headingis "Model"); footer not sticky.CinematicShotList.vue(wasCinematicDirectionGrid.vue): one card, a row per Camera / Shot / Light / Film / Look / Grade (icon or grade swatch, part, choice, chevron), each opening its picker.CinematicStudioPanel.vue: the picker sits in the grid, measures itself and centres on the opening row, clamped to the grid; bottom sheet on mobile. Same 2fr/3fr split asreshoot/ReshootStudio.vue.CinematicSceneField.vue: "View full prompt" andpromptSegmentsremoved (also fromuseCinematicShot); the AI prompt switch sits at the foot of the scene box.CinematicReferenceSlot.vue: empty slots are dashed "+ Character" / "+ Palette" buttons; filled slots show the image.CinematicOutputControls.vue: aspect, resolution and takes as threeCinematicMenus side by side.CinematicDirectionSegments.vue(wasCinematicDirectionChips.vue): Shot, Light, Film, Look, Grade as one segmented control of icons (Grade: its palette swatch); each opens its single-group picker, which closes on a pick. The tabbed direction picker, its Done button andcinematic.picker.doneare removed.CinematicFormatSegments.vue: aspect, resolution and takes (×1–×4) as segmented menus. Options shared with the side panel throughuseFormatMenus.ts.CinematicReferenceMenu.vue: the + opens a References menu (Character, Palette); each item opens its file picker, shows its file once attached, and attached references can be removed there. The references popover and thereferences/format/directionpopover keys are gone (PickerKeyonly).CinematicComposer.vue: the scene grows to four lines then scrolls; the AI prompt switch sits top right (own line on phones); the camera chip shows the body icon and focal length and is named "Camera: ", with the full summary in its tooltip.direction-icons.ts+CinematicDirectionIcon.vue(new): line drawings for every shot, light, film and look option, incurrentColorand site tokens (no hex). Used in the composer segments and shot-list rows instead of preview images;CinematicOptionGridkeeps the preview and adds the icon in a corner. Newdirection-icons.test.ts.CinematicTooltip.vue: one styled reka tooltip for the model menu, camera chip, direction and format segments, aspect toggles and the AI prompt switch.CinematicMenutakestooltipand wraps the whole menu, so the dropdown keeps its own popper anchor.CinematicCheckBadge.vue: a yellow check on the chosen starter shot and the chosen picker tile.camera-icons.ts+CinematicOptionIcon.vue: body, lens, focal and aperture icons are shaded shapes in an oblique view. Colours come from site tokens viacolor-mix; unchosen options are dimmed. Newcamera-icons.test.ts.CinematicMenu.vue: chosen row filled instead of checked;sideprop.catalog.ts: distinct preview frames for the default tiles;gradeGrouptypedDirectionGroup<'grade'>.copy.ts:cinematic.section.shot; zh-CN "Optional" fixed (was "竖屏").Review Focus
git diffbetween feat(website): Cinematic Studio opens on a playground-style side panel #19227's head (cebc807) and this PR's head is empty for every changed file. CodeRabbit's findings there were fixed in b60b0eb, which this includes.CinematicMenuandCinematicReferenceMenunow put the tooltip outsideDropdownMenuRoot. happy-dom does not lay out poppers, so this was checked in Chromium.?demo=1(existinguseCinematicDemoRun) renders sample frames without credits, for design review on previews.reshoot/. feat(website): link each workshop app to its GitHub repository #19313 (GitHub link beside the title) also touchesCinematicStudioPanel.vue, but not the same lines.typecheck:website, oxfmt, oxlint, eslint, fallow, the co-author check, and the workshop unit tests (1013).Screenshots (if applicable)
Preview (needs a signed-in staff account), once deploy-preview passes: