iart-ai/kinetic-typography-skills

kinetic-typography

This skill should be used when the user asks to "animate a headline", "make a kinetic typography video", "do a split-text reveal", "stagger text by character/word/line", "make a lyric or caption video", "animate a variable font weight", "put text on a path"…

Voir la source

Mode d’emploi

  1. Copiez la commande et examinez le paquet avant de l’installer dans votre environnement agent.
  2. Consultez le document original pour les outils, identifiants et environnements requis.
  3. Décrivez les entrées et le livrable souhaité ; comparez la sortie aux sources.
npx skills add https://github.com/iart-ai/kinetic-typography-skills --skill kinetic-typography

Avant de commencer

  • L’installation n’inclut ni abonnement au modèle, ni quota API, ni accès aux fichiers.
  • Étoiles et installations reflètent l’activité, pas un test indépendant de qualité ou de sécurité.
Document Skill original

Rendu depuis le dépôt source en conservant titres, exemples, code, tableaux, liens et images.

Typography in Motion

Animate text expressively: splitting copy into lines/words/characters, staggering reveals, masking, blurring, morphing weight, and running text along a path. Produces runnable code for the web (CSS, GSAP, Framer Motion) and video (Remotion), plus an After Effects expression approach.

When to use

  • Headline / tagline entrances and animated title cards.
  • Lyric videos, caption/subtitle reveals, kinetic-typography pieces.
  • Variable-font animation (weight, width, slant, optical size).
  • Text-on-path, text stroke draw-on, and per-character morph effects.

Core workflow

1. Get static typography right first

Motion cannot rescue bad type. Before animating, lock line-height (1.1-1.2 for display headlines, 1.4-1.6 for body), tracking (tighten display by -1% to -3%, e.g. letter-spacing: -0.02em), alignment, and weight. Then split and animate.

2. Choose split granularity

  • By line — calmest, most premium; best for multi-line paragraphs and elegant headlines.
  • By word — energetic, good for taglines and lyric hits.
  • By character — most kinetic/playful; risk of feeling busy on long copy. Reserve for short strings.

Tools: SplitType or GSAP SplitText on the web; .split('') / .split(' ') plus interpolate in Remotion; motion staggered children in Framer Motion.

Critical accessibility note: splitting into per-character spans destroys the text for screen readers and breaks copy-paste. Always set aria-label on the container with the full string and aria-hidden="true" on the split fragments.

3. Apply a reveal technique

The four workhorse reveals:

  • Mask / clip reveal (most robust) — wrap each line in overflow: hidden; animate the child from translateY(100%) to 0. The text wipes up from a hidden baseline. No blur artifacts, GPU-cheap, the industry-standard headline reveal.
  • Blur-in — animate filter: blur(12px) -> blur(0) with opacity 0 -> 1. Soft, premium, "focus-pull" feel. Costlier to render; keep blur radius modest and prefer short durations.
  • Clip-path wipe — animate clip-path: inset(0 100% 0 0) to inset(0 0 0 0) for a directional reveal without moving the glyphs.
  • Character stagger — split to chars, fade/translate each with a 20-40ms offset.

4. Apply principle-correct timing

Inline essentials (so this skill stands alone):

  • Enter with ease-out: cubic-bezier(0.16, 1, 0.3, 1) (easeOutExpo) feels premium.
  • Per-fragment duration 400-600ms.
  • Stagger: lines 60-100ms, words 40-70ms, characters 20-40ms apart.
  • Cap total reveal near ~800ms; for long strings, reduce per-char offset or switch to word/line granularity.

5. Mask-reveal, copy-paste (vanilla CSS + JS line split)

html
<h1 class="reveal" aria-label="Designed for motion">Designed for motion</h1>
css
.reveal .line { overflow: hidden; }
.reveal .line > span {
  display: inline-block;
  transform: translateY(110%);
  animation: rise 0.6s cubic-bezier(0.16, 1, 0.3, 1) forwards;
}
.reveal .line:nth-child(2) > span { animation-delay: 0.08s; }
.reveal .line:nth-child(3) > span { animation-delay: 0.16s; }
@keyframes rise { to { transform: translateY(0); } }

6. Variable-font weight animation

css
@keyframes thicken {
  from { font-variation-settings: "wght" 200; }
  to   { font-variation-settings: "wght" 800; }
}
.headline {
  font-family: "Inter var", sans-serif;
  animation: thicken 0.8s cubic-bezier(0.65, 0, 0.35, 1) forwards;
}

Animate font-variation-settings (covers any axis: wght, wdth, slnt, opsz, plus custom axes), not font-weight, to hit non-standard values and to interpolate smoothly. Confirm the loaded font is actually a variable font and the axis range (e.g. wght 100-900) before animating, or the value clamps silently.

Deliver & verify (standalone HTML)

Packaged helper (scripts/): scripts/seek-shot.sh anim.html 0 1.5 3 freezes the ?t=N harness and screenshots each moment; scripts/contact-sheet.sh sheet.png frame-*.png tiles them for one-glance review. See scripts/README.md.

For a web text reveal / kinetic-title deliverable, ship one `.html` file that opens directly in a browser — no build step. One file is the right tier for a headline, title card, or single reveal.

Output contract:

  • One .html file: markup with aria-label on the container, the reveal in inline <style> (or one inline GSAP <script> from CDN).
  • One animation driver — a CSS @keyframes reveal, or a single GSAP timeline; not both.
  • Include the freeze harness below so any moment can be screenshotted deterministically.
  • Split AFTER document.fonts.ready so line breaks are correct in the frozen frame.

Freeze harness — pin a frame for screenshots.

html
<script>
  const t = new URLSearchParams(location.search).get("t");
  if (t !== null) {
    // CSS @keyframes reveal:
    document.querySelectorAll(".reveal *").forEach(el => {
      el.style.animationDelay = (-parseFloat(t)) + "s";
      el.style.animationPlayState = "paused";
    });
    // GSAP timeline instead? → tl.pause(); tl.seek(parseFloat(t));
  }
  window.__ready = true;                                          // ready signal for headless wait
</script>

Verify loop — render → freeze → screenshot → check:

  1. Open the file frozen at start / mid / end: …/type.html?t=0, ?t=<dur/2>, ?t=<dur>.
  2. Screenshot each frozen frame.
  3. Check fidelity (matches the brief) and artifacts — clipped glyphs/descenders, masks that don't fully hide the baseline, FOUC or wrong line breaks before fonts load, blur fringing.
bash
npx playwright screenshot --wait-for-timeout=500 "file://$PWD/type.html?t=0.4" frame-mid.png

Before you finish:

  1. Opens standalone in a browser — no console errors, no missing CDN/font.
  2. One reveal driver; ?t=N freezes the exact frame correctly.
  3. Screenshotted at start / mid / end — no clipped glyphs, no FOUC, line breaks correct.
  4. prefers-reduced-motion honored (text shown in final state, no motion).
  5. aria-label on the container; split fragments aria-hidden — copy and screen-reader text intact.

Quick reference

EffectCore CSS / approach
Mask up revealparent overflow:hidden; child translateY(110%)->0
Blur infilter: blur(12px)->0 + opacity
Directional wipeclip-path: inset(0 100% 0 0)->inset(0)
Char staggersplit chars, 20-40ms offset, fade+rise
Variable weightanimate font-variation-settings:"wght" a->b
Text on pathSVG <textPath href="#curve">
Stroke draw-onSVG text as path, animate stroke-dashoffset

Text on a path (SVG)

html
<svg viewBox="0 0 600 200">
  <path id="curve" d="M20,150 Q300,20 580,150" fill="none" />
  <text font-size="42" fill="#fff">
    <textPath href="#curve" startOffset="0%">Follow the curve</textPath>
  </text>
</svg>

Animate the startOffset attribute (0% -> 100%) to scroll text along the path, or animate the path d to make text flow on a morphing curve.

Reference files

  • references/reveal-recipes.md — copy-paste reveal cookbook: mask/blur/clip/char-stagger in CSS, GSAP SplitText timelines, Framer Motion staggered variants, Remotion per-character interpolation, variable-font keyframes, and a screen-reader-safe split helper.

Tâches connexes