@autoaiai/aicut 0.1.11 → 0.1.13

package/skills/brewreel/docs/gallery.en.md

@@ -1,44 +0,0 @@
-# Screenshots
-
-[← Back to README](../README.en.md) · [中文](gallery.md)
-
-
-Everything below is an actual render. The storyboards are all in the repo, and the products and numbers are fictional. The [demo video](https://github.com/Finderchangchang/brewreel/releases/download/v0.3.0/demo-5MB.mp4) (attached to v0.3.0, about 5 MB) shows actual renders of all three recipes.
-
-**Three recipes**
-
-<table align="center">
-<tr><td align="center"><img src="images/style-cards.png" width="760" alt="cards style" /><br/><sub><b>cards</b> (default, 9:16): gradient background, a centered white card and bold outlined captions, one point per shot</sub></td></tr>
-<tr><td align="center"><img src="images/style-quiz.png" width="760" alt="quiz style" /><br/><sub><b>quiz</b> (9:16): circle a common misconception in red pen → ask one multiple-choice question → reveal → a dictionary card explains it</sub></td></tr>
-<tr><td align="center"><img src="images/style-journey.png" width="760" alt="journey style" /><br/><sub><b>journey</b> (4:5 / 9:16): an original mascot crosses a paper-cut city in one take, one postcard per stop for each category</sub></td></tr>
-</table>
-
-**Six industries** (each image shows frame 0, the cover, on the left and one frame from the middle on the right)
-
-<table align="center">
-<tr>
-<td align="center"><img src="images/ledger.png" width="260" alt="Software: budgeting" /><br/><sub>Software · budgeting (ledger)</sub></td>
-<td align="center"><img src="images/food.png" width="260" alt="Food: item launch" /><br/><sub>Food · item launch (food)</sub></td>
-<td align="center"><img src="images/ecommerce.png" width="260" alt="Ecommerce: physical goods" /><br/><sub>Ecommerce · physical goods (ecommerce)</sub></td>
-</tr>
-<tr>
-<td align="center"><img src="images/education.png" width="260" alt="Education: adult vocational course" /><br/><sub>Education · adult vocational course (education)</sub></td>
-<td align="center"><img src="images/beauty.png" width="260" alt="Beauty: no real photos" /><br/><sub>Beauty · no real photos (beauty)</sub></td>
-<td align="center"><img src="images/travel.png" width="260" alt="Travel: lodging" /><br/><sub>Travel · lodging (travel)</sub></td>
-</tr>
-</table>
-
-<details>
-<summary>More samples: relationship chat app, meeting notes, English captions</summary>
-
-<table align="center">
-<tr>
-<td align="center"><img src="images/jev.png" width="260" alt="Software: relationship chat" /><br/><sub>Software · relationship chat (jev)</sub></td>
-<td align="center"><img src="images/meeting.png" width="260" alt="Software: meeting notes" /><br/><sub>Software · meeting notes (meeting)</sub></td>
-<td align="center"><img src="images/en-focus.png" width="260" alt="English captions" /><br/><sub>English captions (en-focus)</sub></td>
-</tr>
-</table>
-
-</details>
-
-These 9 storyboards are in `examples/` (all in the cards recipe); the quiz and journey samples are in `styles/quiz/examples/` and `styles/journey/examples/`. Every one passes validation and renders with `make.mjs` as-is.

package/skills/brewreel/docs/gallery.md

@@ -1,44 +0,0 @@
-# 截图
-
-[← 返回 README](../README.md) · [English](gallery.en.md)
-
-
-下面都是实际渲染的画面。分镜都在仓库里，产品和数据都是虚构的；[演示视频](https://github.com/Finderchangchang/brewreel/releases/download/v0.3.0/demo-5MB.mp4)（v0.3.0 附件，约 5 MB）是三种配方的实际渲染效果。
-
-**三种配方**
-
-<table align="center">
-<tr><td align="center"><img src="images/style-cards.png" width="760" alt="cards 卡片信息流" /><br/><sub><b>cards 卡片信息流</b>（默认，9:16）：渐变底 + 居中白卡片 + 描边大字幕，一镜讲一件事</sub></td></tr>
-<tr><td align="center"><img src="images/style-quiz.png" width="760" alt="quiz 答题互动" /><br/><sub><b>quiz 答题互动</b>（9:16）：红笔圈出一个常见误解 → 出一道选择题 → 揭晓 → 词条卡讲清楚</sub></td></tr>
-<tr><td align="center"><img src="images/style-journey.png" width="760" alt="journey 角色漫游" /><br/><sub><b>journey 角色漫游</b>（4:5 / 9:16）：原创吉祥物一镜到底横穿剪纸城市，每站一张明信片讲一个类别</sub></td></tr>
-</table>
-
-**六个行业**（每张图左边是第 0 帧封面，右边是片中一帧）
-
-<table align="center">
-<tr>
-<td align="center"><img src="images/ledger.png" width="260" alt="软件：记账工具" /><br/><sub>软件 · 记账工具（ledger）</sub></td>
-<td align="center"><img src="images/food.png" width="260" alt="餐饮：单品上新" /><br/><sub>餐饮 · 单品上新（food）</sub></td>
-<td align="center"><img src="images/ecommerce.png" width="260" alt="电商：实物商品" /><br/><sub>电商 · 实物商品（ecommerce）</sub></td>
-</tr>
-<tr>
-<td align="center"><img src="images/education.png" width="260" alt="教培：成人职业课程" /><br/><sub>教培 · 成人职业课程（education）</sub></td>
-<td align="center"><img src="images/beauty.png" width="260" alt="美业：无实拍照片" /><br/><sub>美业 · 无实拍照片（beauty）</sub></td>
-<td align="center"><img src="images/travel.png" width="260" alt="文旅：住宿" /><br/><sub>文旅 · 住宿（travel）</sub></td>
-</tr>
-</table>
-
-<details>
-<summary>更多样例：情感聊天 App、会议纪要、英文字幕</summary>
-
-<table align="center">
-<tr>
-<td align="center"><img src="images/jev.png" width="260" alt="软件：情感聊天" /><br/><sub>软件 · 情感聊天（jev）</sub></td>
-<td align="center"><img src="images/meeting.png" width="260" alt="软件：会议纪要" /><br/><sub>软件 · 会议纪要（meeting）</sub></td>
-<td align="center"><img src="images/en-focus.png" width="260" alt="英文字幕" /><br/><sub>英文字幕（en-focus）</sub></td>
-</tr>
-</table>
-
-</details>
-
-这 9 份分镜在 `examples/`（都是 cards 配方），quiz 和 journey 的样例在 `styles/quiz/examples/`、`styles/journey/examples/`。每份都能直接通过校验并用 `make.mjs` 出片。

package/skills/brewreel/docs/how-it-works.en.md

@@ -1,64 +0,0 @@
-# How it works
-
-[← Back to README](../README.en.md) · [中文](how-it-works.md)
-
-
-```
-Product brief
-  → pick a recipe: meta.style = cards / quiz / journey, meta.industry = an industry pack
-  → the cheap model writes storyboard.json (pick shots, fill in text; no code, no coordinates)
-  → validate.mjs (hard rules + ad-law / industry compliance; the report goes back to the model)
-  → brew with make.mjs (music → render → contact sheet → check frames → delivery manifest)
-  → human pre-publish check ("Pre-publish checklist" in SKILL.md)
-```
-
-- **Brief**: the merchant, or you, fills in a product brief; the generic template is [`brief-template.md`](../brief-template.md), and each industry has its own `industries/<id>/brief-template.en.md`.
-- **Pick a recipe**: choose the recipe and industry pack by content and put them in the storyboard's `meta`.
-- **Write the storyboard**: the model reads `SKILL.md` (the instructions for AI assistants; English version `SKILL.en.md`), the recipe's `recipes.md` and the industry's `recipe.md`, and outputs one JSON file.
-- **Validate**: `node scripts/validate.mjs storyboard.json` produces a report that names the shot and field to fix and how; `llm_make.py` and the plugin feed the report back to the model verbatim.
-- **Brew**: `make.mjs` renders in one command and refuses delivery if any self-check has a ✗; see ["Render QA and the delivery manifest"](features.en.md#render-qa-and-the-delivery-manifest).
-- **Human check**: delivery comes with a pre-publish checklist of things a machine can't judge (photo authorization, genuine reviews, price conditions and so on) for the publisher to confirm.
-
-<details>
-<summary><b>Directory structure</b></summary>
-
-```
-brewreel/
-  SKILL.md / SKILL.en.md      The instructions an AI assistant reads (how to pick shots, fill fields, follow the flow)
-  README.md / README.en.md    The project front page for humans
-  LICENSE / NOTICE / THIRD_PARTY_LICENSES.md
-  shots.md / shots.en.md      Parameter reference for all 19 shots
-  brief-template.md           Generic brief template
-  docs/
-    *.md / *.en.md            README detail pages: install, features, screenshots, FAQ, how it works, known limitations
-    shots/                    Detailed per-shot docs (19 shots × 2 languages)
-    images/                   The logo and example images used in the READMEs
-  industries/                 The six industry packs
-  scripts/
-    validate.mjs              Validation entry point
-    make.mjs                  One-command render
-    llm_make.py               Headless mode: brief → storyboard → video
-    make_bgm.py               Parametric original background music
-    build_docs.mjs            Generates shots.md / shots.en.md from spec.json files
-    privacy-scan.mjs          Pre-publish privacy self-check
-    check-originality.mjs     Originality check: color difference from the reference (CIEDE2000) + signature-by-signature record
-    checks/ lib/              Rule implementations
-  template/                   The Remotion rendering project
-    src/shots/                19 shared shot components + parameter specs (used by cards)
-    src/styles/               Per-recipe manifest, design tokens and shots; registry.gen.ts is generated by scripts/gen-styles.mjs
-    src/core/                 Fonts, themes, animation, layout helpers
-    src/illust/               Industry fallback illustrations
-    public/                   Fonts, sound effects, sample assets
-  examples/                   9 ready-to-render storyboard samples (six industries + English + two generic samples, all cards)
-  styles/                     Recipes (style packs): nine-layer spec, narrative templates, rules, examples; _template/ scaffolds a new style
-  distill/                    Recipe crafting and reusable prompts (breakdown → replicate → redesign → componentize → test → review)
-  CONTRIBUTING.en.md          How to contribute (use / remix / create, PR checklist)
-  integrations/
-    deepseek-harness/         DeepSeek Harness plugin dsh-brewreel (7 tools)
-  tests/
-    validate/                 Positive/negative regression tests for the validator
-    rules/                    Regression tests for each industry's rules (4 cases each)
-    originality/              Unit tests for the originality check (color difference, rules, signature records)
-```
-
-</details>

package/skills/brewreel/docs/how-it-works.md

@@ -1,64 +0,0 @@
-# 它怎么工作
-
-[← 返回 README](../README.md) · [English](how-it-works.en.md)
-
-
-```
-产品简报（brief）
-  → 选配方：meta.style 选 cards / quiz / journey，meta.industry 选行业包
-  → 便宜模型写分镜 storyboard.json（挑镜头、填文字，不写代码、不写坐标）
-  → 校验 validate.mjs（硬性规则 + 广告法 / 行业合规，报错回喂给模型改）
-  → 开酿 make.mjs（配乐 → 渲染 → 拼图 → 检查帧 → 交付清单）
-  → 发布前人工自查（SKILL.md 的「发布前自查清单」）
-```
-
-- **简报**：商家或你自己填一份产品简报，通用模板是 [`brief-template.md`](../brief-template.md)，各行业另有 `industries/<id>/brief-template.md`。
-- **选配方**：按内容挑配方和行业包，写进分镜的 `meta`。
-- **写分镜**：模型读 `SKILL.md`（AI 助手的说明书）、配方的 `recipes.md` 和行业的 `recipe.md`，只输出一份 JSON。
-- **校验**：`node scripts/validate.mjs storyboard.json` 出中文报告，指出哪一镜哪个字段要改、怎么改；`llm_make.py` 和插件会把报告原样回喂给模型。
-- **开酿**：`make.mjs` 一条命令出片，任何一项自查有 ✗ 都不交付，见[「出片质检与交付清单」](features.md#出片质检与交付清单)。
-- **人工自查**：交付时附一份发布前自查清单，列出机器判断不了的事（照片授权、评价真实性、价格条件等），由发布者确认。
-
-<details>
-<summary><b>目录结构</b></summary>
-
-```
-brewreel/
-  SKILL.md / SKILL.en.md      AI 助手看的说明书（怎么挑镜头、写字段、走流程）
-  README.md / README.en.md    人看的项目首页
-  LICENSE / NOTICE / THIRD_PARTY_LICENSES.md
-  shots.md / shots.en.md      19 个镜头的参数总表
-  brief-template.md           通用简报模板
-  docs/
-    *.md / *.en.md            README 的详细页：安装、功能、截图、常见问题、原理、已知限制
-    shots/                    每个镜头的详细文档（19 × 2 语言）
-    images/                   README 里的 logo 和效果示例图
-  industries/                 六个行业包
-  scripts/
-    validate.mjs              校验入口
-    make.mjs                  一条命令出片
-    llm_make.py               无头模式：简报 → 分镜 → 出片
-    make_bgm.py               参数化原创配乐
-    build_docs.mjs            从 spec.json 生成 shots.md / shots.en.md
-    privacy-scan.mjs          发布前隐私自查
-    check-originality.mjs     原创性检查：和参考片的配色色差（CIEDE2000）+ 招牌逐条记录
-    checks/ lib/              校验规则的实现
-  template/                   Remotion 渲染工程
-    src/shots/                19 个公共镜头组件 + 参数 spec（cards 配方用）
-    src/styles/               各配方的清单、设计令牌、专属镜头；registry.gen.ts 由 scripts/gen-styles.mjs 生成
-    src/core/                 字体、主题、动画、版式等公共逻辑
-    src/illust/               行业插画兜底
-    public/                   字体、音效、示例素材
-  examples/                   9 份可直接渲染的分镜样例（六行业 + 英文 + 两份通用示例，都是 cards 配方）
-  styles/                     配方（风格包）：九层规格、叙事模板、规则、样例；_template/ 是新风格脚手架
-  distill/                    调配方流程和可复用提示词（拆解 → 复刻 → 再设计 → 组件化 → 测试 → 评审）
-  CONTRIBUTING.md             贡献说明（一创 / 二创 / 三创、PR 自查清单）
-  integrations/
-    deepseek-harness/         DeepSeek Harness 插件 dsh-brewreel（7 个工具）
-  tests/
-    validate/                 校验规则的正负例回归测试
-    rules/                    六个行业的规则回归测试（各 4 例）
-    originality/              原创性检查（色差算法、判定规则、招牌记录）的单元测试
-```
-
-</details>

package/skills/caption-animation/SKILL.md

@@ -1,216 +0,0 @@
----
-name: caption-animation
-description: "This skill should be used when the user asks to \"add animated captions\", \"make TikTok captions\", \"karaoke captions\", \"word-by-word subtitles\", \"auto subtitles\", \"highlight the active word\", \"burn in captions to a video\", \"sync captions to a voiceover/narration\", or \"turn an SRT/transcript into animated text\". Covers word-level timing from Whisper, per-word pop/scale-in, active-word highlight, 9:16 safe-area placement, readable type, and burn-in vs sidecar SRT/VTT. Runs in AICUT: the film is film.html, a page per shot, drawn by `aicut look` and `aicut render`."
-license: MIT
-source: "iart-ai/tiktok-video-skills"
----
-
-# Caption Animation
-
-## Running it in AICUT
-
-AICUT's film is `film.html`: it cuts pages, footage, pictures and sound together on tracks. A page is a web page that
-sets `window.film = { frame(t) }` and draws that moment — no framework, no build step, no composition object, no
-`registerRoot`. Five shots are five pages cut together, not one page that changes scene, so each shot stays trimmable
-and replaceable in Studio.
-
-```sh
-npx -y @autoaiai/aicut@latest          # the manual: the format and the commands, always current
-npx -y @autoaiai/aicut@latest open .   # make this folder a film and print Studio's address
-npx -y @autoaiai/aicut@latest look     # a contact sheet of the whole film, its sound, and a determinism check
-npx -y @autoaiai/aicut@latest look 4.5 # one frame, full size — look at the PNG, not at the code
-npx -y @autoaiai/aicut@latest render   # the MP4, beside film.html
-```
-
-Run the manual before the first film: it is authoritative for the format, and this skill is about the method above it.
-
-Where the words below name another program's part, it becomes this:
-
-| elsewhere | in AICUT |
-|---|---|
-| composition, timeline, project file | `film.html`; tracks and clips inside it |
-| a scene, a sequence, a slide | a page (`pages/<name>.html`) cut onto a track |
-| `useCurrentFrame()`, `interpolate()` | `frame(t)` — the picture depends only on `t`, never on state or `Date.now()` |
-| a still, a screengrab | `aicut look <t>` |
-| `render` / `build` / `export` | `aicut render` (the person can also export from Studio) |
-| a dev server, a preview window | `aicut open` and the Studio address it prints |
-
-Hard rules that hold whatever the method says: the picture depends only on `t`; every value in the person's prompt is
-binding; look at the frames `look` writes before saying a film is done.
-
-
-Turn a transcript or voiceover into word-timed, scroll-stopping captions: each word pops in on the syllable, the active word highlights, and the text stays glued to the narration. Build the kind of captions that drive watch-time on TikTok, Reels, and Shorts — readable, on-beat, and inside the safe area.
-
-## When to use
-
-- Word-by-word or karaoke captions on short-form vertical video (9:16).
-- Burning open captions onto a voiceover/narration track from a transcript.
-- Turning a Whisper/SRT transcript into animated text in Remotion or on the web.
-- Restyling auto-generated subtitles into a branded, "Hormozi-style" reveal.
-
-## The pipeline
-
-| Stage | Job | Tool |
-|---|---|---|
-| Transcribe | Audio → word-level timestamps | Whisper (`@remotion/install-whisper-cpp`), AssemblyAI |
-| Normalize | → `{ text, startMs, endMs }[]` tokens | `@remotion/captions` `Caption` type |
-| Page | Group words into 1–4 word "pages" | `createTikTokStyleCaptions()` |
-| Animate | Per-word pop + active-word highlight | Remotion `spring()` / CSS |
-| Place | Safe-area, readable type | layout rules below |
-| Export | Burn-in (MP4) and/or sidecar SRT/VTT | Remotion render / file emit |
-
-## Two non-negotiable rules
-
-1. **Word-level timing, not line-level.** Karaoke reads as magic only when each word lands on the syllable. Always transcribe to word timestamps; never fake them by splitting a line evenly over its duration — drift is instantly visible.
-2. **Readability beats style.** Captions are read on a phone, in sunlight, muted. Bold sans-serif, heavy stroke or shadow, high contrast first; decoration second. A caption nobody can read is decoration, not a caption.
-
-## Word timing from a transcript
-
-Whisper emits word-level timestamps directly. Normalize every source into one flat token shape so the renderer never cares where the words came from:
-
-```ts
-// Caption token — the one shape everything downstream consumes
-type Token = { text: string; startMs: number; endMs: number };
-
-// Parse an SRT cue block "00:00:01,200 --> 00:00:01,640" into ms
-const toMs = (t: string) => {
-  const [h, m, rest] = t.split(":");
-  const [s, ms] = rest.split(",");
-  return ((+h * 60 + +m) * 60 + +s) * 1000 + +ms;
-};
-```
-If only sentence-level cues exist, re-transcribe for word timing — do not interpolate. Whisper's `medium.en` is ~2x faster than `large` and accurate enough for clean voice audio.
-```
-
-## Per-word pop-in (Remotion)
-
-Drive each word's scale and opacity off a spring anchored to its own `startMs`. The micro-overshoot is what makes it feel "alive".
-
-```tsx
-import { useCurrentFrame, useVideoConfig, spring, interpolate } from "remotion";
-
-const Word: React.FC<{ token: Token; active: boolean }> = ({ token, active }) => {
-  const frame = useCurrentFrame();
-  const { fps } = useVideoConfig();
-  const enter = (token.startMs / 1000) * fps;            // word's own entrance frame
-  const p = spring({ frame: frame - enter, fps, config: { damping: 12, mass: 0.6 } });
-  const scale = interpolate(p, [0, 1], [0.6, 1]);        // pop from 60% → 100% (overshoots)
-  return (
-    <span style={{
-      display: "inline-block",
-      transform: `scale(${scale})`,
-      opacity: interpolate(p, [0, 1], [0, 1]),
-      color: active ? "#FFE45E" : "#FFFFFF",             // active-word highlight
-      transition: "color 80ms linear",
-    }}>{token.text}&nbsp;</span>
-  );
-};
-```
-Compute `active` by testing `frame` against each token's `[startMs, endMs]`. Keep the highlight a single accent color or a filled "box" behind the live word — never animate every word's color at once.
-```
-
-## Pages, not walls of text
-
-Show 1–4 words at a time. Use Remotion's `createTikTokStyleCaptions()` (or group manually) and tune `combineTokensWithinMilliseconds`: ~200–500ms for true word-by-word, ~1000–1500ms for short readable phrases.
-
-```ts
-import { createTikTokStyleCaptions } from "@remotion/captions";
-const { pages } = createTikTokStyleCaptions({
-  captions, combineTokensWithinMilliseconds: 1200,       // ~2–3 words per page
-});
-
-## Web/CSS pop (no Remotion)
-
-For a DOM/live player, give each word its own `animation-delay` equal to `startMs`:
-
-```css
-.word { display:inline-block; opacity:0; animation: pop .26s cubic-bezier(.34,1.56,.64,1) forwards; }
-@keyframes pop { from { opacity:0; transform:translateY(.18em) scale(.7) } to { opacity:1; transform:none } }
-.word.active { color:#FFE45E; }
-```js
-words.forEach(w => { const el = mk(w.text); el.style.animationDelay = `${w.startMs}ms`; track.append(el); });
-
-## Readable type and safe-area placement (9:16, 1080×1920)
-
-| Property | Value | Why |
-|---|---|---|
-| Font | Bold/ExtraBold sans (Montserrat, Inter, Helvetica) | Reads on small, busy screens |
-| Size | 56–80px (≈8% of frame height), min 45px | Legible muted on a phone |
-| Stroke | 2–6px solid black outline | Survives any background |
-| Shadow | Soft drop shadow as backup to stroke | Separation on bright frames |
-| Case | Uppercase or sentence; high contrast fill | Punch + scannability |
-| Vertical pos | Center band, ~62–70% down | Above the UI, below the action |
-| Bottom safe | Keep clear of bottom ~280px / 15% | Avoids caption/CTA/audio UI |
-| Side safe | Keep within center 80% width | Avoids right-rail icons |
-
-Place captions in the center band — not pinned to the very bottom, where the platform UI (username, audio, buttons) lives. Pad text to a max of ~2 lines.
-
-## Sync to a voiceover / narration
-
-The transcript already carries the timing of the exact audio. Lay that same audio under the composition and keep timestamps in the audio's timebase — captions and voice stay locked with zero manual nudging. If the VO is re-recorded, re-transcribe; never hand-shift offsets.
-
-## Burn-in vs sidecar SRT/VTT
-
-| | Burn-in (open) | Sidecar SRT/VTT (closed) |
-|---|---|---|
-| Where | Pixels in the MP4 | Separate `.srt`/`.vtt` file |
-| Social (TikTok/Reels) | Required — guaranteed, styleable | Often ignored by the platform |
-| Accessibility (ADA/WCAG) | Does not satisfy alone | Required (toggleable) |
-| Best practice | Burn animated captions for social | Also ship a sidecar for the web/SEO |
-
-Emit both: burn the animated captions for social, and write a plain SRT/VTT from the same tokens for accessible/SEO playback.
-
-## Output checklist
-
-- Word-level timestamps (real, from transcription) — no even-split fakery.
-- Per-word pop with micro-overshoot; one accent color for the active word.
-- 1–4 words per page; never a wall of text.
-- Bold sans, 56–80px, 2–6px stroke + shadow, high contrast.
-- Center band, clear of bottom ~280px and side rails.
-- Audio laid under composition; timestamps in the audio timebase.
-- Ship burn-in for social and a sidecar SRT/VTT for accessibility.
-
-## Deliver & verify (rendered stills → MP4)
-
-> **Packaged helper** (`scripts/`): tile your stills with `scripts/contact-sheet.sh sheet.png f-hook.png f-mid.png f-end.png`, then assert the encode with `scripts/probe-mp4.sh out.mp4 [WxH] [fps]`. See `scripts/README.md`.
-
-Captions ship as a Remotion composition (`<Composition>` + zod `schema` + `defaultProps`) — all word motion frame-driven off `useCurrentFrame()`, never `Date.now()` / `Math.random()` / timers. Deliverable = `out/*.mp4` (burned-in) + the project + the sidecar SRT/VTT. 9:16 vertical (1080×1920) is the default.
-
-**Verify loop — render stills → inspect → encode.** Word timing is the thing that breaks; check it at exact frames before you spend an encode.
-
-# Stills at start / a sampled active-word frame / end — WITH SHIPPED PROPS (real tokens + audio)
-npx remotion still Captions out/f-start.png --frame=0   --props='{"captionsSrc":"vo.json"}'
-npx remotion still Captions out/f-mid.png   --frame=90  --props='{"captionsSrc":"vo.json"}'
-npx remotion still Captions out/f-end.png   --frame=N   --props='{"captionsSrc":"vo.json"}'  # N = durationInFrames-1
-
-# Inspect each PNG:
-
-#  - the word highlighted at frame 90 is the word whose [startMs,endMs] contains 90/fps (no drift)
-
-#  - burn-in legible: bold sans, stroke+shadow holds, no clipping
-
-#  - 9:16 safe area: caption sits in the center band, clear of top ~12% and bottom ~20-35% (captions/CTA/audio UI) and the right action rail
-
-npx remotion render Captions out/captions.mp4 --props='{"captionsSrc":"vo.json"}'   # encode once stills are right
-npx remotion render Captions out/demo.gif --codec=gif                                # README proof clip
-```
-Use `npx remotion compositions` to read `durationInFrames`/`fps` and pick the active-word + end frames.
-
-**Before you finish:**
-1. Stills render cleanly at frame 0, a mid active-word frame, and last — no missing font/audio.
-2. The correct word is highlighted at the sampled frame (frame/fps lands inside its token); no even-split fakery.
-3. Burn-in is legible (stroke+shadow) and the caption is fully inside the 9:16 safe area at every checked frame.
-4. Frame-driven only — no `Date.now()` / `Math.random()` / timers.
-5. Shipped props (real tokens, not just `defaultProps`) render correctly; MP4 + sidecar SRT/VTT emitted, GIF optional.
-```
-
-## Reference files
-
-- `references/word-timed-captions.md` — end-to-end build: Whisper transcription and the `Caption` type, a full SRT parser, Remotion word-timed component with active-highlight, manual paging, an SRT/VTT emitter, per-platform safe-area maps, and a readable-type spec sheet.
-
-## Source
-
-Rewritten for AICUT from [`iart-ai/tiktok-video-skills`](https://github.com/iart-ai/tiktok-video-skills)
-(MIT), skills/caption-animation/SKILL.md. The method is that skill's own; the way it runs is AICUT's.
-
-The repository holds 4 skills; this is the one its entry stands for.