Check default path Skill

Codex compatibility note:

Invoke repository skills with $skill-name in Codex; this mirrored copy rewrites legacy Claude /skill-name references.

Task tracker mandate: BEFORE executing any workflow or skill step, create/update task tracking for all steps and keep it synchronized as progress changes.

User-question prompts mean to ask the user directly in Codex.

Ignore Claude-specific mode-switch instructions when they appear.

Strict execution contract: when a user explicitly invokes a skill, execute that skill protocol as written.

Subagent authorization: when a skill is user-invoked or AI-detected and its protocol requires subagents, that skill activation authorizes use of the required spawn_agent subagent(s) for that task.

Do not skip, reorder, or merge protocol steps unless the user explicitly approves the deviation first.

For workflow skills, execute each listed child-skill step explicitly and report step-by-step evidence.

If a required step/tool cannot run in this environment, stop and ask the user before adapting.

Codex Project-Reference Loading (No Hooks)

Codex uses static project-reference loading instead of runtime-injected project docs. When coding, planning, debugging, testing, or reviewing, open project docs explicitly using this routing.

Always read:

docs/project-config.json (project-specific paths, commands, modules, and workflow/test settings)
docs/project-reference/docs-index-reference.md (routes to the full docs/project-reference/* catalog)
docs/project-reference/lessons.md (always-on guardrails and anti-patterns)

Missing/stale context route: If docs/project-config.json, the docs index, lessons.md, CLAUDE.md, AGENTS.md, or any task-required reference doc is missing or stale, auto-run $project-init or the narrow setup route ($project-config, $docs-init, $scan-all, $scan --target=<key>, $claude-md-init) before ordinary project-specific work. If Codex mirrors or AGENTS.md are missing/stale, ask the user to run $sync-codex; do not auto-run it.

Situation-based docs:

Backend/CQRS/API/domain/entity changes: backend-patterns-reference.md, domain-entities-reference.md, project-structure-reference.md
Frontend/UI/styling/design-system: frontend-patterns-reference.md, scss-styling-guide.md, design-system/README.md
Spec authoring, docs/specs/ pathing, or TC format: feature-spec-reference.md, spec-system-reference.md, spec-principles.md
Behavior/public-contract changes or spec-test-code sync: workflow-spec-test-code-cycle-reference.md plus the spec docs above
Derived spec indexes/ERDs/reimplementation guides: spec-system-reference.md and source Feature Specs under docs/specs/
Integration test implementation/review: integration-test-reference.md
E2E test implementation/review: e2e-test-reference.md
Code review/audit work: code-review-rules.md plus domain docs above based on changed files

Do not read all docs blindly. Start from docs-index-reference.md, then open only relevant files for the task.

[IMPORTANT] Use task tracking to break ALL work into small tasks BEFORE starting. For simple tasks, ask user whether to skip. MUST ATTENTION wait for user approval of scene plan (Step 3.3) before writing any files — NEVER skip. MUST ATTENTION read existing project files before modifying — NEVER overwrite scenes blindly. MUST ATTENTION update totalChapters in ALL existing scene files when adding/removing scenes.

Be skeptical. Apply critical thinking. Every claim needs traced proof, confidence >80% to act.

Quick Summary

Goal: Create new Remotion video project, add/update scenes in existing one, or launch local Remotion Studio preview server.

Two Modes:

| Mode | When | Action | | ----------------------------- | ------------------------------------------------------------------- | ---------------------------------------------- | | Create / Update (default) | User describes video content OR asks to update/add scenes | Scaffold new project or modify existing scenes | | Play | User says "play", "preview", "open studio", "watch", "start server" | Find Remotion project → npm run studio |

Default project path: remotion/ (relative to workspace root). Respect any path user explicitly provides.

Key Rules:

NEVER implement scenes before user approves the plan (Step 3.3)
Always read existing project structure before adding/modifying scenes
Keep totalChapters consistent across ALL scene files
All animations MUST be driven by useCurrentFrame() — CSS transitions FORBIDDEN
Always use Remotion components (<Img>, <Video>, <Audio>) instead of native HTML elements

Implementing one of these? Copy from refs/ — do NOT implement from memory:

| Implementing... | Copy from | | ---------------------------------------------------- | ------------------------------------------------------------------------------------------------------------- | | {source-root}/components/Shared.tsx (new scaffold) | refs/Shared.tsx — C palette, ProgressBar, ChapterBadge, CodeBlock, Pill, AnimRow | | {source-root}/utils/animations.ts (new scaffold) | refs/animations.ts — easeOut, easeInOut, pop, staggeredEaseOut, counter | | Typewriter or word-highlight text effect | refs/text-animations.tsx — getTypedText, Cursor, TypewriterScene, Highlight | | TikTok-style captions with word highlight | refs/captions.tsx — CaptionedVideo, CaptionPage, delayRender, createTikTokStyleCaptions | | Video/audio duration, dimensions, frames | refs/mediabunny-utils.ts — getVideoDuration, getAudioDuration, getVideoDimensions, canDecode, extractFrames | | Mapbox map scene | refs/maps-mapbox.tsx — MapScene component, interactive:false, camera animation | | ElevenLabs TTS voiceover generation | refs/generate-voiceover.ts — TTS script + calculateMetadata integration |

Phase 0: Mode Detection

Classify prompt:

PLAY keywords   → "play", "preview", "open studio", "start server", "watch", "launch", "dev server"
CREATE/UPDATE   → everything else (default)

Ambiguous → proceed with Create/Update (safe default).

Phase 1A: Detect Existing Project

Locate Remotion project before acting:

# Check default path
ls remotion/package.json 2>/dev/null

# Check if CWD is already a Remotion project
ls package.json 2>/dev/null | xargs grep -l '"remotion"' 2>/dev/null

# Check common paths
find . -maxdepth 3 -name "package.json" -exec grep -l '"remotion"' {} \; 2>/dev/null | head -5

State after detection:

PROJECT_EXISTS = true/false
PROJECT_PATH = remotion/ (default) or detected path
HAS_STUDIO_SCRIPT = true/false (check package.json scripts.studio)

Phase 2: PLAY Mode

When: User explicitly asks to preview / open studio / launch dev server.

Step 2.1 — Verify project exists

If PROJECT_EXISTS = false:

"No Remotion project found. Run $remotion <description> to create one first." Exit.

Step 2.2 — Find launch command

cat {PROJECT_PATH}/package.json | grep '"studio"'
# → use: npm run studio
# → fallback: npx remotion studio

Step 2.3 — Start dev server (background)

cd {PROJECT_PATH} && npm run studio

Remotion Studio launches at http://localhost:3000

Server runs in background. Report URL and composition IDs visible in {source-root}/Root.tsx.

Step 2.4 — Optional: one-frame render check

npx remotion still [composition-id] --scale=0.25 --frame=30
# At 30fps, --frame=30 = one-second mark (zero-based)

Phase 3: CREATE / UPDATE Mode

Step 3.1 — New vs Update Decision

| Condition | Action | | ------------------------ | ------------------------------------- | | PROJECT_EXISTS = false | Scaffold new project → Phase 3.2 | | PROJECT_EXISTS = true | Read existing project → Phase 3.4 |

Step 3.2 — Scaffold New Project

Only when no existing Remotion project found.

3.2.1 Bootstrap with create-video (preferred)

# Creates {PROJECT_PATH}/ with blank template (no Tailwind)
npx create-video@latest --yes --blank --no-tailwind {PROJECT_PATH}
cd {PROJECT_PATH}
npm install @remotion/transitions  # add transitions support

Replace generated {source-root}/ with project structure below (keep package.json and tsconfig.json from scaffold).

Fallback (manual) — when `npx create-video` unavailable

mkdir -p {PROJECT_PATH}/{source-root}/compositions {PROJECT_PATH}/{source-root}/components {PROJECT_PATH}/{source-root}/utils
cd {PROJECT_PATH}
npm init -y
npm install remotion @remotion/cli @remotion/transitions react react-dom
npm install -D @types/react @types/react-dom typescript

Directory structure (both paths)

{PROJECT_PATH}/
  package.json          ← scripts: studio, render, still
  tsconfig.json
  {source-root}/
    index.ts            ← registerRoot
    Root.tsx            ← register compositions
    components/
      Shared.tsx        ← palette + reusable UI components
    utils/
      animations.ts     ← easeOut, staggeredEaseOut, pop, counter
    compositions/
      {CompositionName}.tsx

3.2.3 Create `package.json` scripts (merge into existing after install)

{
    "scripts": {
        "studio": "remotion studio",
        "render": "remotion render {CompositionId} out/video.mp4",
        "still": "remotion still {CompositionId} --frame=0 out/still.png"
    },
    "remotion": {
        "entryPoint": "{source-root}/index.ts"
    }
}

3.2.4 Create `tsconfig.json`

{
    "compilerOptions": {
        "target": "ES2020",
        "lib": ["dom", "ES2020"],
        "jsx": "react-jsx",
        "module": "commonjs",
        "moduleResolution": "node",
        "strict": true,
        "esModuleInterop": true,
        "skipLibCheck": true,
        "outDir": "dist"
    },
    "include": ["src"]
}

3.2.5 Create `{source-root}/index.ts`

import { registerRoot } from 'remotion';
import { Root } from './Root';
registerRoot(Root);

3.2.6 Create `{source-root}/components/Shared.tsx`

Copy from refs/Shared.tsx — do NOT implement from memory. Exports: C (palette), ProgressBar, ChapterBadge, CodeBlock, Pill, AnimRow. Always create; all scenes import from here.

3.2.7 Create `{source-root}/utils/animations.ts`

Copy from refs/animations.ts — do NOT implement from memory. Exports: easeOut, easeInOut, pop, staggeredEaseOut, counter.

Step 3.3 — Plan Composition Structure

MUST ATTENTION wait for user confirmation before implementing any scenes.

Plan composition before writing files:

Parse intent — What video about? What information to convey?
Decide scene count — 1 scene per major concept (30s → ~4 scenes, 60s → ~8, 90s → ~12)
Assign durations — Each scene: 6–12s (180–360 frames @ 30fps)
Name scenes — Scene01Intro, Scene02{Topic}, etc.
Write scene brief — what shown, key data, visual layout (left/right split, full-width, grid)

Present plan before writing files:

Proposed: {N} scenes, ~{total}s total
  Scene 01 ({Xs}) — {topic}: {layout description}
  Scene 02 ({Xs}) — {topic}: {layout description}
  …
Proceed? (or adjust)

Step 3.4 — Read Existing Project (UPDATE path)

MUST ATTENTION read existing project state before any modifications.

When PROJECT_EXISTS = true:

# Read composition registry
cat {PROJECT_PATH}/{source-root}/Root.tsx

# List scene files
ls {PROJECT_PATH}/{source-root}/compositions/ 2>/dev/null || ls {PROJECT_PATH}/{source-root}/scenes/ 2>/dev/null

# Read main composition orchestrator
cat {PROJECT_PATH}/{source-root}/ClaudeAgentExplainer.tsx 2>/dev/null  # or equivalent

Identify:

Existing scene count and names
Current totalChapters / composition duration
Where to insert / which scenes to modify

Target only scenes affected by user's request. Preserve all unchanged scenes.

Step 3.5 — Implement Scene Files

Scene file anatomy (follow exactly)

import { AbsoluteFill, useCurrentFrame, interpolate } from 'remotion';
import { C, ProgressBar, ChapterBadge } from '../components/Shared';  // adjust path
import { easeOut, staggeredEaseOut } from '../utils/animations';       // adjust path

// Data arrays at the top — keep out of component body
const ITEMS = [ ... ];

export const Scene{NN}{Name}: React.FC = () => {
    const frame = useCurrentFrame();

    return (
        <AbsoluteFill style={{ background: C.bg, fontFamily: 'system-ui, -apple-system, sans-serif' }}>
            <ChapterBadge index={NN} label="{Scene Name}" color={C.blue} />

            <div style={{ position: 'absolute', inset: 0, display: 'flex', gap: 48, padding: '68px 72px 44px' }}>
                {/* Content here */}
            </div>

            <ProgressBar chapterIndex={NN - 1} totalChapters={TOTAL} />
        </AbsoluteFill>
    );
};

Animation rules

| Element | Pattern | | --------------------- | --------------------------------------------------------- | | Eyebrow label | opacity: easeOut(frame, 0, 14) | | Hero title | opacity: easeOut(frame, 8, 22) + translateY from 28→0 | | Subtitle | opacity: easeOut(frame, 22, 18) | | List items (stagger) | staggeredEaseOut(frame, i, startAt, 12, 16) | | Cards (stagger up) | staggeredEaseOut + translateY(20→0) | | Cards (stagger right) | staggeredEaseOut + translateX(28→0) | | Late callout boxes | easeOut(frame, 90+, 18) |

Layout patterns

Left/Right split (most common):

<div style={{ position: 'absolute', inset: 0, display: 'flex', gap: 48, padding: '68px 72px 44px' }}>
    <div style={{ width: 400, flexShrink: 0, ... }}>  {/* Left panel */}
    <div style={{ flex: 1, ... }}>                     {/* Right panel */}
</div>

Full-width column:

<div style={{ position: 'absolute', inset: 0, display: 'flex', flexDirection: 'column', padding: '68px 64px 44px', gap: 16 }}>

Centered (title/CTA scene):

<div style={{ position: 'absolute', inset: 0, display: 'flex', flexDirection: 'column', justifyContent: 'center', alignItems: 'center', padding: '72px 100px', textAlign: 'center' }}>

Card grid:

<div style={{ display: 'grid', gridTemplateColumns: 'repeat(4, 1fr)', gap: 12 }}>

Color and font guidelines

| Element | fontSize | fontWeight | Notes | | ---------------- | -------- | ---------- | ----------------------------------------------------------------- | | Eyebrow label | 14 | 700 | category color, letterSpacing: 3 | | Hero title | 44–56 | 800 | C.text, lineHeight: 1.1 | | Body text | 17–21 | 400 | C.dim, lineHeight: 1.6 | | Card label | 15–16 | 700 | category color | | Monospace detail | 12–14 | 400 | fontFamily: 'Courier New' | | Card container | — | — | C.surface, borderLeft: 3px solid color, borderRadius: 10–12 |

Step 3.6 — Wire Root.tsx

Multi-scene composition (with transitions)

import { TransitionSeries, linearTiming } from '@remotion/transitions';
import { fade } from '@remotion/transitions/fade';
import { Scene01Intro } from './compositions/Scene01Intro';
// ... other imports

const T = 18; // transition duration in frames

const D = {
    s01: 210, // 7s
    s02: 240 // 8s
    // ...
};

const SCENES = Object.values(D).length;
export const TOTAL_DURATION_FRAMES = Object.values(D).reduce((a, b) => a + b, 0) - (SCENES - 1) * T;

const tr = () => <TransitionSeries.Transition presentation={fade()} timing={linearTiming({ durationInFrames: T })} />;

export const {
    CompositionName
}: React.FC = () => (
    <TransitionSeries>
        <TransitionSeries.Sequence durationInFrames={D.s01}>
            <Scene01Intro />
        </TransitionSeries.Sequence>
        {tr()}
        <TransitionSeries.Sequence durationInFrames={D.s02}>
            <Scene02Next />
        </TransitionSeries.Sequence>
        {/* ... */}
    </TransitionSeries>
);

// Root.tsx
import { Composition } from 'remotion';
import { {CompositionName}, TOTAL_DURATION_FRAMES } from './{CompositionName}';

export const Root: React.FC = () => (
    <Composition id="{CompositionId}" component={{{CompositionName}}} durationInFrames={TOTAL_DURATION_FRAMES} fps={30} width={1920} height={1080} />
);

Single-scene composition

// Root.tsx
import { Composition } from 'remotion';
import { MyScene } from './compositions/MyScene';

export const Root: React.FC = () => <Composition id="MyVideo" component={MyScene} durationInFrames={300} fps={30} width={1920} height={1080} />;

Step 3.7 — Post-scaffold: launch prompt

After creating/updating, report changed files and offer launch:

✅ Created {N} scene files in {PROJECT_PATH}/{source-root}/compositions/
   Total duration: ~{X}s ({FRAMES} frames @ 30fps)

To preview: run `$remotion play` — starts Remotion Studio at http://localhost:3000
To render:  cd {PROJECT_PATH} && npm run render

Update Mode: Specific Scenarios

Add a new scene

Read Root.tsx to find composition orchestrator and current scene count
Read totalChapters across existing scenes
Determine insertion point (end = safest)
Create new scene file following anatomy above
Update Root / composition orchestrator:
- Add import
- Add D.sNN entry
- Add <TransitionSeries.Sequence> block
- MUST ATTENTION update totalChapters in ALL existing scene files (+1)

Modify an existing scene

Read specific scene file
Identify data array or JSX block needing change
Make surgical edits only — do not touch unrelated sections

Change visual style / palette

Edit {source-root}/components/Shared.tsx → C object
Font changes: update fontFamily in AbsoluteFill style per scene (or add global in Shared)

Remotion API Reference

Authoritative Remotion API reference — apply when implementing beyond basic scene creation.

Animations (Core Rules)

MUST ATTENTION ALL animations driven by useCurrentFrame(). Durations: seconds × fps.

import { useCurrentFrame, useVideoConfig, interpolate, Easing } from 'remotion';

export const FadeIn = () => {
    const frame = useCurrentFrame();
    const { fps } = useVideoConfig();
    const opacity = interpolate(frame, [0, 2 * fps], [0, 1], {
        extrapolateRight: 'clamp',
        extrapolateLeft: 'clamp',
        easing: Easing.bezier(0.16, 1, 0.3, 1)
    });
    return <div style={{ opacity }}>Hello World!</div>;
};

FORBIDDEN: CSS transition-*, CSS animation-*, Tailwind transition-*/animate-* classes.

Timing — interpolate & Bézier Easing

// With clamping + Bézier easing
const opacity = interpolate(frame, [0, 60], [0, 1], {
    easing: Easing.bezier(0.16, 1, 0.3, 1),
    extrapolateLeft: 'clamp',
    extrapolateRight: 'clamp'
});

Copy-paste curves:

easing: Easing.bezier(0.16, 1, 0.3, 1); // crisp UI entrance (strong ease-out)
easing: Easing.bezier(0.45, 0, 0.55, 1); // editorial / slow fade
easing: Easing.bezier(0.34, 1.56, 0.64, 1); // playful overshoot

Composing multiple properties from one progress:

const slideIn = interpolate(frame, [start, start + dur], [0, 1], {
    easing: Easing.bezier(0.22, 1, 0.36, 1),
    extrapolateLeft: 'clamp',
    extrapolateRight: 'clamp'
});
const overlayX = interpolate(slideIn, [0, 1], [100, 0]);
const opacity = interpolate(slideIn, [0, 1], [0, 1]);

Sequencing & Trimming

<Sequence> delays element appearance. useCurrentFrame() inside = local frame (0-based). Always premountFor:

<Sequence from={1 * fps} durationInFrames={2 * fps} premountFor={1 * fps}>
    <Title />
</Sequence>

<Series> for sequential without overlap:

<Series>
    <Series.Sequence durationInFrames={45}>
        <Intro />
    </Series.Sequence>
    <Series.Sequence durationInFrames={60}>
        <MainContent />
    </Series.Sequence>
    <Series.Sequence offset={-15} durationInFrames={60}>
        <SceneB />
    </Series.Sequence>{' '}
    {/* overlap */}
</Series>

Trimming:

<Sequence from={-0.5 * fps}><MyAnimation /></Sequence>       // skip first 15 frames
<Sequence durationInFrames={1.5 * fps}><MyAnimation /></Sequence>  // trim end
<Sequence from={30}><Sequence from={-15}><MyAnimation /></Sequence></Sequence>  // trim + delay

Compositions

import { Composition, Folder, Still } from "remotion";

<Composition id="MyComposition" component={MyComp} durationInFrames={100} fps={30} width={1080} height={1080} />

// With default props (use `type` not `interface`)
<Composition id="MyComposition" component={MyComp} durationInFrames={100} fps={30} width={1080} height={1080}
  defaultProps={{ title: "Hello World", color: "#ff0000" } satisfies MyCompositionProps}
/>

// Still (single-frame)
<Still id="Thumbnail" component={Thumbnail} width={1280} height={720} />

Assets — staticFile() & public folder

MUST use staticFile() for assets in public/:

import { Img, staticFile } from "remotion";
import { Video } from "@remotion/media";
import { Audio } from "@remotion/media";

<Img src={staticFile("logo.png")} />
<Video src={staticFile("clip.mp4")} />
<Audio src={staticFile("music.mp3")} />
// Remote URLs work directly — no staticFile needed

Images

MUST use <Img> from remotion — NOT <img>, NOT Next.js <Image>, NOT CSS background-image.

import { Img, staticFile } from "remotion";

<Img src={staticFile("photo.png")} style={{ width: 500, height: 300, objectFit: "cover" }} />
<Img src={staticFile(`frames/frame${frame}.png`)} />  // dynamic paths work

Videos

Install: npx remotion add @remotion/media

import { Video } from "@remotion/media";

<Video src={staticFile("video.mp4")} />
<Video src={staticFile("video.mp4")} trimBefore={2 * fps} trimAfter={10 * fps} />
<Video src={staticFile("video.mp4")} volume={(f) => interpolate(f, [0, fps], [0, 1], { extrapolateRight: "clamp" })} />
<Video src={staticFile("video.mp4")} playbackRate={2} loop muted />

Audio

Install: npx remotion add @remotion/media

import { Audio } from "@remotion/media";

<Audio src={staticFile("audio.mp3")} />
<Audio src={staticFile("audio.mp3")} trimBefore={2 * fps} trimAfter={10 * fps} />
<Sequence from={1 * fps}><Audio src={staticFile("audio.mp3")} /></Sequence>  // delay
// volume callback: `f` starts at 0 when audio begins, not composition frame
<Audio src={staticFile("audio.mp3")} volume={(f) => interpolate(f, [0, fps], [0, 1], { extrapolateRight: "clamp" })} />

GIFs & Animated Images

import { AnimatedImage, staticFile } from "remotion";

// Preferred: supports GIF, APNG, AVIF, WebP
<AnimatedImage src={staticFile("animation.gif")} width={500} height={500} />
<AnimatedImage src={staticFile("animation.gif")} width={500} height={300} fit="cover" playbackRate={2} />

// Fallback: GIF only — npx remotion add @remotion/gif
import { Gif } from "@remotion/gif";
<Gif src={staticFile("animation.gif")} width={500} height={500} />

Fonts

Google Fonts (recommended): npx remotion add @remotion/google-fonts

import { loadFont } from '@remotion/google-fonts/Roboto';

const { fontFamily } = loadFont('normal', { weights: ['400', '700'], subsets: ['latin'] });
<h1 style={{ fontFamily, fontSize: 80, fontWeight: 'bold' }}>{text}</h1>;

Local fonts: npx remotion add @remotion/fonts

import { loadFont } from '@remotion/fonts';
import { staticFile } from 'remotion';

await Promise.all([
    loadFont({ family: 'Inter', url: staticFile('Inter-Regular.woff2'), weight: '400' }),
    loadFont({ family: 'Inter', url: staticFile('Inter-Bold.woff2'), weight: '700' })
]);

Transitions (Full API)

npx remotion add @remotion/transitions

import { TransitionSeries, linearTiming, springTiming } from "@remotion/transitions";
import { fade } from "@remotion/transitions/fade";
import { slide } from "@remotion/transitions/slide";
import { wipe } from "@remotion/transitions/wipe";
import { flip } from "@remotion/transitions/flip";

<TransitionSeries>
  <TransitionSeries.Sequence durationInFrames={60}><SceneA /></TransitionSeries.Sequence>
  <TransitionSeries.Transition presentation={fade()} timing={linearTiming({ durationInFrames: 15 })} />
  <TransitionSeries.Sequence durationInFrames={60}><SceneB /></TransitionSeries.Sequence>
</TransitionSeries>

// Slide directions: "from-left" | "from-right" | "from-top" | "from-bottom"
<TransitionSeries.Transition presentation={slide({ direction: "from-left" })} timing={linearTiming({ durationInFrames: 20 })} />

// Spring timing
springTiming({ config: { damping: 200 }, durationInFrames: 25 })

// Overlay (no duration shortening — note: cannot be adjacent to transition or another overlay)
<TransitionSeries.Overlay durationInFrames={20} offset={0}><MyEffect /></TransitionSeries.Overlay>

Duration: two 60-frame scenes + 15-frame transition = 60 + 60 - 15 = 105 frames total.

Light Leaks (Overlay Effect)

Requires Remotion ≥ 4.0.415. Check: npx remotion versions. Upgrade: npx remotion upgrade.

npx remotion add @remotion/light-leaks

import { LightLeak } from "@remotion/light-leaks";

// As TransitionSeries overlay
<TransitionSeries.Overlay durationInFrames={30}><LightLeak /></TransitionSeries.Overlay>

// Standalone decorative overlay
<AbsoluteFill>
  <MyContent />
  <LightLeak durationInFrames={60} seed={3} />
</AbsoluteFill>

// seed: different patterns; hueShift: 0=yellow-orange, 120=green, 240=blue
<LightLeak seed={5} hueShift={240} />

3D with Three.js (React Three Fiber)

npx remotion add @remotion/three

MUST wrap in <ThreeCanvas> with width and height props. FORBIDDEN: useFrame() from @react-three/fiber — causes flickering. Use useCurrentFrame().

import { ThreeCanvas } from '@remotion/three';
import { useVideoConfig, useCurrentFrame } from 'remotion';

const { width, height } = useVideoConfig();
const frame = useCurrentFrame();

<ThreeCanvas width={width} height={height}>
    <ambientLight intensity={0.4} />
    <directionalLight position={[5, 5, 5]} intensity={0.8} />
    <mesh rotation={[0, frame * 0.02, 0]}>
        <boxGeometry args={[2, 2, 2]} />
        <meshStandardMaterial color="#4a9eff" />
    </mesh>
</ThreeCanvas>;

// <Sequence> inside ThreeCanvas must use layout="none"

Text Animations

All text animations driven by useCurrentFrame(). CSS transitions FORBIDDEN.

Typewriter — use string slicing, NEVER per-character opacity. Spring word highlight — scaleX wipe from left.

Implementing typewriter or word-highlight? Copy from refs/text-animations.tsx — do NOT implement from memory. Contains: getTypedText, Cursor, TypewriterScene, Highlight.

Text Measurement

npx remotion add @remotion/layout-utils

import { measureText, fitText, fillTextBox } from '@remotion/layout-utils';

const { width, height } = measureText({ text: 'Hello World', fontFamily: 'Arial', fontSize: 32, fontWeight: 'bold' });
const { fontSize } = fitText({ text: 'Hello World', withinWidth: 600, fontFamily: 'Inter', fontWeight: 'bold' });

MUST ATTENTION call only after fonts loaded. Use validateFontIsLoaded: true. Match font properties exactly between measurement and rendering.

DOM Node Measurement

Remotion applies scale() transform — use useCurrentScale() to get true dimensions:

import { useCurrentScale } from 'remotion';

const scale = useCurrentScale();
const rect = ref.current.getBoundingClientRect();
const trueWidth = rect.width / scale;

Captions & Subtitles

npx remotion add @remotion/captions

Caption type: { text, startMs, endMs, timestampMs, confidence }

Transcribe with Whisper.cpp: npx remotion add @remotion/install-whisper-cpp

import { installWhisperCpp, downloadWhisperModel, transcribe, toCaptions } from '@remotion/install-whisper-cpp';

await installWhisperCpp({ to: './whisper.cpp', version: '1.5.5' });
await downloadWhisperModel({ model: 'medium.en', folder: './whisper.cpp' });
// Convert to 16KHz wav first: npx remotion ffmpeg -i input.mp4 -ar 16000 output.wav
const whisperOutput = await transcribe({
    model: 'medium.en',
    whisperPath: './whisper.cpp',
    whisperCppVersion: '1.5.5',
    inputPath: '/path/to/audio.wav',
    tokenLevelTimestamps: true
});
const { captions } = toCaptions({ whisperCppOutput: whisperOutput });

Import from.srt:

import { parseSrt } from '@remotion/captions';
const { captions } = parseSrt({ input: await fetch(staticFile('subtitles.srt')).then(r => r.text()) });

TikTok-style word highlight display:

Implementing TikTok-style captions? Copy from refs/captions.tsx — do NOT implement from memory. Contains: CaptionedVideo, CaptionPage, delayRender, createTikTokStyleCaptions, per-token active highlighting.

Note: use whiteSpace: "pre" — captions are whitespace-sensitive.

Dynamic Compositions — calculateMetadata

import { CalculateMetadataFunction } from 'remotion';

const calculateMetadata: CalculateMetadataFunction<Props> = async ({ props, abortSignal }) => {
    const durationInSeconds = await getVideoDuration(props.videoSrc);
    return { durationInFrames: Math.ceil(durationInSeconds * 30) };
};

// Transform props (fetch data before render)
const calculateMetadata: CalculateMetadataFunction<Props> = async ({ props, abortSignal }) => {
    const data = await fetch(props.dataUrl, { signal: abortSignal }).then(r => r.json());
    return { props: { ...props, fetchedData: data } };
};

// Return values (all optional): durationInFrames, width, height, fps, props, defaultOutName, defaultCodec

Parameters — Zod Schema

import { z } from 'zod';
import { zColor } from '@remotion/zod-types';

export const MyCompositionSchema = z.object({
    title: z.string(),
    color: zColor() // renders color picker in Studio
});

// Root.tsx
<Composition
    id="MyComposition"
    component={MyComponent}
    schema={MyCompositionSchema}
    durationInFrames={100}
    fps={30}
    width={1080}
    height={1080}
    defaultProps={{ title: 'Hello World', color: '#ff0000' }}
/>;

Top-level type MUST be z.object().

Mediabunny — Video/Audio Metadata

npx remotion add mediabunny

Need video/audio duration, dimensions, or frame extraction? Copy from refs/mediabunny-utils.ts — do NOT implement from memory. Contains: getVideoDuration, getAudioDuration, getVideoDimensions, canDecode, extractFrames.

Use staticFile() for local files. Use FileSource instead of UrlSource in Node.js/Bun.

FFmpeg in Remotion

npx remotion ffmpeg -i input.mp4 output.mp3
npx remotion ffprobe input.mp4

Trimming — prefer <Video> props (non-destructive, no re-encoding):

<Video src={staticFile('video.mp4')} trimBefore={5 * fps} trimAfter={10 * fps} />

// FFmpeg fallback (MUST re-encode to avoid frozen frames)
// npx remotion ffmpeg -ss 00:00:05 -i public/input.mp4 -to 00:00:10 -c:v libx264 -c:a aac public/output.mp4

Silence Detection

# Step 1: measure loudness
npx remotion ffmpeg -i public/video.mov -map 0:a -af loudnorm=print_format=json -f null /dev/null
# → read input_i (integrated loudness dB) and input_thresh

# Step 2: detect silences (d=0.5 = min silence duration in seconds)
npx remotion ffmpeg -i public/video.mov -map 0:a -af "silencedetect=noise=${THRESH}dB:d=0.5" -f null /dev/null

<Video src={staticFile('video.mov')} trimBefore={Math.floor(leadingEnd * fps)} trimAfter={Math.ceil(trailingStart * fps)} />

Audio Visualization

npx remotion add @remotion/media-utils

import { useWindowedAudioData, visualizeAudio, visualizeAudioWaveform, createSmoothSvgPath } from '@remotion/media-utils';

const { audioData, dataOffsetInSeconds } = useWindowedAudioData({ src: staticFile('music.mp3'), frame, fps, windowInSeconds: 30 });
if (!audioData) return null;

// Spectrum bars (numberOfSamples must be power of 2)
const frequencies = visualizeAudio({ fps, frame, audioData, numberOfSamples: 256, optimizeFor: 'speed', dataOffsetInSeconds });
// Values 0-1; left=bass, right=highs

// Waveform SVG
const waveform = visualizeAudioWaveform({ fps, frame, audioData, numberOfSamples: 256, windowInSeconds: 0.5, dataOffsetInSeconds });
const path = createSmoothSvgPath({ points: waveform.map((y, i) => ({ x: (i / (waveform.length - 1)) * width, y: 100 + y * 100 })) });

// Bass-reactive scale
const bassIntensity = frequencies.slice(0, 32).reduce((sum, v) => sum + v, 0) / 32;
const scale = 1 + bassIntensity * 0.5;

MUST ATTENTION pass frame from parent to child visualization — NEVER call useCurrentFrame() in each child inside <Sequence>.

Logarithmic scaling: const db = 20 * Math.log10(value); const scaled = (db - (-100)) / ((-30) - (-100));

Lottie Animations

npx remotion add @remotion/lottie

import { Lottie, LottieAnimationData } from '@remotion/lottie';
import { cancelRender, continueRender, delayRender } from 'remotion';

export const MyAnimation = () => {
    const [handle] = useState(() => delayRender('Loading Lottie animation'));
    const [animationData, setAnimationData] = useState<LottieAnimationData | null>(null);

    useEffect(() => {
        fetch('https://assets4.lottiefiles.com/packages/lf20_zyquagfl.json')
            .then(r => r.json())
            .then(json => {
                setAnimationData(json);
                continueRender(handle);
            })
            .catch(err => {
                cancelRender(err);
            });
    }, [handle]);

    if (!animationData) return null;
    return <Lottie animationData={animationData} style={{ width: 400, height: 400 }} />;
};

Charts & Data Visualization

MUST ATTENTION drive all animations from useCurrentFrame(). Disable all third-party chart library animations.

import { spring, evolvePath, getLength, getPointAtLength, getTangentAtLength } from '@remotion/paths';

// Bar chart — spring stagger
const bars = data.map((item, i) => {
    const height = spring({ frame, fps, delay: i * 5, config: { damping: 200 } });
    return <div style={{ height: height * item.value }} />;
});

// Line chart path animation (npx remotion add @remotion/paths)
const pathProgress = interpolate(frame, [0, 2 * fps], [0, 1], { extrapolateLeft: 'clamp', extrapolateRight: 'clamp' });
const { strokeDasharray, strokeDashoffset } = evolvePath(pathProgress, svgPath);
<path d={svgPath} fill="none" stroke="#FF3232" strokeWidth={4} strokeDasharray={strokeDasharray} strokeDashoffset={strokeDashoffset} />;

Maps with Mapbox

npm i mapbox-gl @turf/turf @types/mapbox-gl — set REMOTION_MAPBOX_TOKEN in .env.

Implementing Mapbox map scene? Copy from refs/maps-mapbox.tsx — do NOT implement from memory. Key rules: interactive: false, fadeDuration: 0, explicit width/height/position: "absolute" on container, animate camera via useCurrentFrame(), NO cleanup (_map.remove()). Render: npx remotion render --gl=angle --concurrency=1.

Transparent Video Rendering

# ProRes (for video editors)
npx remotion render --image-format=png --pixel-format=yuva444p10le --codec=prores --prores-profile=4444 MyComp out.mov

# WebM VP9 (for browsers)
npx remotion render --image-format=png --pixel-format=yuva420p --codec=vp9 MyComp out.webm

AI Voiceover (ElevenLabs TTS)

Implementing ElevenLabs TTS voiceover? Copy from refs/generate-voiceover.ts — do NOT implement from memory. Includes calculateMetadata integration to size composition to audio duration.

Sound Effects

import { Audio } from '@remotion/sfx';
<Audio src={'https://remotion.media/whoosh.wav'} />;

Available: whoosh, whip, page-turn, switch, mouse-click, shutter-modern, shutter-old, ding, bruh, vine-boom, windows-xp-error — all at https://remotion.media/{name}.wav

TailwindCSS

Use if installed. NEVER use transition-* or animate-* Tailwind classes — animate via useCurrentFrame().

Critical Thinking Mindset — Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination.

AI Mistake Prevention — Failure modes to avoid on every task:

Re-read files after context changes. Context compaction, resume, or long-running work can make memory stale; verify current files before acting. Verify generated content against source evidence. AI hallucinates APIs, names, claims, and document facts. Check the relevant source before documenting or referencing. Check downstream references before deleting or renaming. Removing an artifact can stale docs, generated mirrors, configs, and callers; map references first. Trace the full impact chain after edits. Changing a definition can miss derived outputs and consumers. Follow the affected chain before declaring done. Verify ALL affected outputs, not just the first. One green check is not all green checks; validate every output surface the change can affect. Assume existing values are intentional — ask WHY before changing. Before changing a constant, limit, flag, wording, or pattern, read nearby context and history. Surface ambiguity before acting — don't pick silently. Multiple valid interpretations require an explicit question or stated assumption with risk. Keep shared guidance role-relevant. Universal guidance must help every receiving skill or agent; code-specific obligations belong only in code-specific protocols.

Closing Reminders

Protocols in force (concise digest of the SYNC/shared blocks this skill carries):

Critical Thinking: MUST ATTENTION apply critical + sequential thinking; every claim needs traced proof, confidence >80% to act.
AI Mistake Prevention: verify generated content against evidence, trace downstream references, verify all affected outputs, re-read after context loss, surface ambiguity.
MUST ATTENTION wait for user approval of scene plan (Step 3.3) — NEVER implement scenes before approval
MUST ATTENTION read existing project structure (Step 3.4) before modifications — NEVER overwrite blindly
MUST ATTENTION update totalChapters in ALL scene files when adding/removing scenes — one missed file causes visual regression
MUST ATTENTION keep data arrays outside component body — NEVER define const ITEMS inside component function
MUST ATTENTION use npx create-video@latest --yes --blank --no-tailwind for scaffold — NEVER npm init unless fallback needed
MUST ATTENTION use staggeredEaseOut for list/card reveals — NEVER all-at-once opacity
MUST ATTENTION verify PROJECT_EXISTS before Play mode — report missing project and exit
MUST ATTENTION use task tracking to plan ALL work before starting — mark each task done immediately
MUST ATTENTION ALL animations driven by useCurrentFrame() — CSS transitions, CSS animations, Tailwind animate/transition classes FORBIDDEN
MUST ATTENTION use Remotion components <Img>, <Video>, <Audio> — NEVER native HTML elements
MUST ATTENTION use staticFile() for all public/ folder assets — NEVER raw relative paths
MUST ATTENTION Copy from appropriate refs/ file — NEVER implement text animations, captions, mediabunny, maps, or voiceover from memory

MUST ATTENTION apply critical + sequential thinking — every claim needs appropriate traced evidence (file:line for repo/code claims; source URL or artifact section for research, product, content, and docs claims); confidence >80% to act, <60% DO NOT recommend. Anti-hallucination: never present guess as fact, admit uncertainty freely, cross-reference independently, stay skeptical of own confidence.

MUST ATTENTION apply AI mistake prevention — verify generated content against evidence, trace downstream references before deleting or renaming, verify all affected outputs, re-read files after context loss, and surface ambiguity before acting.

Hookless Prompt Protocol Mirror (Auto-Synced)

Source: .claude/.ck.json + .claude/skills/shared/sync-inline-versions.md (:full blocks) + .claude/scripts/lib/hookless-prompt-protocol.cjs

[WORKFLOW-EXECUTION-PROTOCOL] [BLOCKING] Workflow Execution Protocol — MANDATORY IMPORTANT MUST CRITICAL. Do not skip for any reason.

Generic portability boundary: Reusable skills and protocol text stay project-neutral; project-specific conventions are discovered from docs/project-config.json and docs/project-reference/. Apply shared AI-SDD from shared/sdd-artifact-contract.md. Read docs/project-config.json and docs/project-reference/docs-index-reference.md, then open the project reference docs named there. For spec, test-case, behavior-change, public-contract, or docs/specs/ work, route through the local spec docs named by the docs index: feature-spec-reference.md, spec-system-reference.md, spec-principles.md, and workflow-spec-test-code-cycle-reference.md when specs/tests/code must stay synchronized. If either file or a required reference doc is missing or stale, auto-run $project-init (or the narrow lower-level route such as $project-config, $docs-init, $scan-all, or $scan --target=<key>) before ordinary project-specific work. Any supported AI tool may execute when this shared context and local docs are available.

DETECT: If the prompt starts with an explicit slash skill/workflow command, execute it directly. Otherwise match the prompt against the workflow catalog and skill list.
ANALYZE: Choose the best option: execute directly, invoke a skill, activate a standard workflow, or compose a custom step combination.
AUTO-SELECT: Pick the best option yourself. Do not ask the user to choose between direct execution, skill, standard workflow, or custom workflow.
ACTIVATE: For a selected workflow, call $start-workflow <workflowId>; for a selected skill, invoke that skill; for a custom workflow, sequence custom steps directly; for direct execution, proceed with the task.
CREATE TASKS: task tracking for ALL workflow/skill/custom steps before execution when the selected path has multiple steps.
EXECUTE: Advance per the Workflow Step Advancement & Parallel Phases rule in your context instructions — model-driven; a sub-agent completion advances a step identically to an inline call; a parallel-phase group is an all-return barrier (advance only after ALL members return, never serialize it)

Shared AI-SDD Protocol Markers

Source: .claude/skills/shared/sync-inline-versions.md

SYNC:ai-sdd-artifact-contract

AI-SDD Artifact Contract — Shared spec-driven development rules stay portable and source-owned.

Keep reusable AI-SDD principles in .claude; put repository-specific paths, commands, owners, products, and formats in project config/reference docs.

Preserve cycle: spec -> plan -> tasks -> implement -> verify -> update spec/docs.

Trace every requirement or invariant through decision, task, TC/test, source evidence, and docs/spec update.

Treat code-to-spec extraction as reference-only until accepted by the canonical spec owner.

Any supported AI tool may plan, implement, review, or verify with synced context; using multiple tools is optional.

Update .claude source first, then sync generated mirrors; do not manually edit .agents, .codex, or AGENTS.md. — why: mirrors are generated artifacts; hand-edits are overwritten on the next sync

If docs/project-config.json, root instruction files, or a required project-reference doc is missing or stale, auto-run $project-init or the narrow lower-level route before ordinary project-specific work.

Active reference: shared/sdd-artifact-contract.md in the active skills root.

SYNC:ai-sdd-artifact-contract:reminder

MANDATORY Apply shared/sdd-artifact-contract.md; keep reusable AI-SDD in .claude and local rules in project docs.
MANDATORY Code-to-spec extraction is reference-only until canonical acceptance; any supported AI tool may execute with synced context.
MANDATORY Update .claude source before syncing generated mirrors; do not manually edit .agents, .codex, or AGENTS.md.
MANDATORY Missing or stale project config, root instruction files, or required reference docs route project-specific work through $project-init or the narrow setup route automatically. [TASK-PLANNING] [MANDATORY] BEFORE executing any workflow or skill step, create/update task tracking for all planned steps, then keep it synchronized as each step starts/completes.

[LESSON-LEARNED-REMINDER] [BLOCKING] Task Planning & Continuous Improvement — MANDATORY. Do not skip.

Break work into small tasks (task tracking) before starting. Add final task: "Analyze AI mistakes & lessons learned".

Extract lessons — ROOT CAUSE ONLY, not symptom fixes:

Name the FAILURE MODE (reasoning/assumption failure), not symptom — "assumed API existed without reading source" not "used wrong enum value".
Generality test: does this failure mode apply to ≥3 contexts/codebases? If not, abstract one level up.
Write as a universal rule — strip project-specific names/paths/classes. Useful on any codebase.
Consolidate: multiple mistakes sharing one failure mode → ONE lesson.
Recurrence gate: "Would this recur in future session WITHOUT this reminder?" — No → skip $learn.
Auto-fix gate: "Could $code-review/$code-simplifier/$security-review/$lint catch this?" — Yes → improve review skill instead.
BOTH gates pass → ask user to run $learn. [CRITICAL-THINKING-MINDSET] Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence >80% to act. Anti-hallucination principle: Never present guess as fact — cite sources for every claim, admit uncertainty freely, self-check output for errors, cross-reference independently, stay skeptical of own confidence — certainty without evidence root of all hallucination. AI Attention principle (Primacy-Recency): Put the 3 most critical rules at both top and bottom of long prompts/protocols so instruction adherence survives long context windows. Goal-driven execution: Define success criteria first, loop until verified, and stop only when observable checks pass. Tests verify intent: Tests must protect business rules/invariants and fail when the protected intent breaks, not only mirror current behavior.

Common AI Mistake Prevention (System Lessons)

Re-read files after context compaction. Edit requires prior Read in same context; compaction wipes read state. Re-read before editing.
Grep for old terms after bulk replacements. AI over-trusts find/replace completeness. Grep full repo after bulk edits for missed refs in docs/configs/catalogs.
Check downstream references before deleting. Deletions cascade doc/code staleness. Map referencing files before removal.
After memory loss, check existing state before creating new. Compaction wipes prior-work memory. Query current state to resume — never blindly duplicate.
Verify AI-generated content against actual code. AI hallucinates APIs, class names, method signatures. Grep to confirm existence before documenting/referencing.
Trace full dependency chain after edits. Changing a definition misses downstream consumers. Trace the full chain.
When renaming, grep ALL consumer file types. Some file types silently ignore missing refs (no compile error). Search code, templates, configs, generated files.
Trace ALL code paths when verifying correctness. Code existing ≠ code executing. Trace early exits, error branches, conditional skips — not just happy path.
Update docs that embed canonical data when source changes. Docs inlining derived data (workflows, schemas, configs) go stale silently. Update all embedding docs alongside source.
Verify sub-agent results after context recovery. Background agents may finish while parent compacted — grep-verify output, don't trust assumed completion.
Cross-check full target list against sub-agent assignments. Parallel sub-agents by category miss boundary items. Reconcile union of assignments against target list before proceeding.
Sub-agents inherit knowledge only from their agent .md definition — use custom agent types, not built-in Explore. Tool adoption = permission + knowledge + enforcement (numbered workflow step).
Persist sub-agent findings incrementally, not as a final batch. Long sub-agents hit cutoffs before final write — findings lost. Instruct append-per-section to report file.
When debugging, ask "whose responsibility?" before fixing. Trace caller (wrong data) vs callee (wrong handling). Fix at responsible layer — never patch symptom site.
Grep ALL removed names after extraction/refactoring. Primary file "done" ≠ secondary files clean. Grep entire scope for every removed symbol before declaring complete.
Assume existing values are intentional — ask WHY before changing. Pattern-matching as "wrong" skips context. Before changing any constant/limit/flag: read comments, git blame, surrounding code.
Verify ALL affected outputs, not just the first. One build green ≠ all green. Multi-stack changes (backend/frontend/tests/docs) require verifying EVERY output.
Evaluate fit before copying a nearby pattern. Closest example ≠ matching preconditions — verify the new context shares the same constraints, base classes, scope, lifetime.
Holistic-first debugging — resist nearest-attention trap. Don't dive into first plausible cause. List EVERY precondition (config, env vars, paths, DB, endpoints, creds, versions, DI, data). Verify each against evidence (grep/query — not reasoning). Ask "what would falsify this?" — if nothing, it's not a hypothesis. Most expensive failure: going deeper in "obvious" layer while bug sits in layer never questioned.
Surgical changes — apply the diff test (context-aware). Two modes: (1) Bug fix → every line traces to the bug; no restyling; orphan cleanup only for imports YOUR changes made unused. (2) Review/enhancement → implement improvements AND announce as "Enhancement beyond main request: [what]". Never silently scope-creep. Diff test: "Would this line exist if I wasn't asked to do X?" — if no, delete or announce.
Surface ambiguity before coding — don't pick silently. Multiple valid interpretations → present each with effort: "[Request] could mean (1) [N h], (2) [N h]. Which matters?" List scope/format/volume/constraints assumptions first. If simpler path exists, say so. Never silently pick.
[MANDATORY FIRST ACTION] ALWAYS activate a suitable skill or workflow BEFORE responding. Match task against workflow catalog + skill list; invoke via skill invocation or $start-workflow <workflowId>. NEVER answer or write code before checking. Skip = protocol violation.
Why-Review adversarial mindset — apply when reviewing any plan, decision, or design. Default SKEPTIC not VALIDATOR: steel-man a rejected alternative, invert each stated reason ("what does it sacrifice?"), stress-test top 2-3 assumptions, run pre-mortem ("ships, fails in 3 months — what breaks?"), surface 1-2 alternatives author missed. Section presence ≠ quality; quality = causal reasoning + concrete mitigations + evidence, not "it's better" or "monitor closely".
Front-load report-write in sub-agent prompts for large reviews. Many-file sub-agents hit budget before final write — findings lost. Design prompts so: (1) report-write is first explicit deliverable, (2) append per-file/section (not batched), (3) scope bounded so reads don't exhaust budget. Truncated mid-sentence with no report file → spawn narrower scope, don't retry same prompt.
After context compaction, re-verify all prior phase outcomes before continuing. Summaries describe intent, not environment state (git index, filesystem, processes). On resume, FIRST audit: git status, re-read modified files, verify filesystem. Every "completed" claim is an untested hypothesis until evidence confirms.
OOM/memory: check row count before row size. Triage: (1) Unbounded query — no DB filter for trigger? Push filter to DB; eliminates OOM. (2) Large rows? Projection reduces proportionally. Row reduction > projection in ROI.
Keep domain concepts out of generic/shared/infrastructure layers. Reusable layer (shared library, framework, infra module) must reference NO consumer-specific domain concept — tenant/customer/product IDs, business entities, feature rules. Leak compiles + runs → passes review silently while coupling the "reusable" layer to one consumer. Keep shared type domain-free; push domain fields/logic down into the consumer via subclass/composition. — why: a layer coupled to one consumer's domain is no longer reusable.

Agent Skills: Check default path

Install this agent skill to your local

Skill Files