Whiteboard Animation logo

Whiteboard Animation

Organization
iart-ai
whiteboard-animation

This skill should be used when the user asks to "make a whiteboard animation", "draw-on explainer", "VideoScribe style", "hand-drawn explainer video", "animate a sketch being drawn", "do a doodle/scribe video", "show a hand drawing the illustration", or "animate handwriting appearing on a whiteboard". Covers SVG stroke draw-on, a marker hand that follows the path tip, staggered reveal, erase/wipe transitions, and narrated export.

Overview

Publisheriart-ai
Repositoryexplainer-video-skills
Skill namewhiteboard-animation
Stars
73
Forks
14
Bundled files
2
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 2 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by iart-ai on GitHub. Read the source before you install it.

Installation

Install the Whiteboard Animation AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/iart-ai/explainer-video-skills/tree/main/skills/whiteboard-animation
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/iart-ai/explainer-video-skills.git /tmp/explainer-video-skills
mkdir -p .claude/skills
cp -r /tmp/explainer-video-skills/skills/whiteboard-animation .claude/skills/whiteboard-animation
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Whiteboard Animation in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Whiteboard Animation on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Whiteboard Animation is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Whiteboard Animation

The VideoScribe / Doodly look: a hand draws illustrations and handwriting onto a board, stroke by stroke, while narration explains. Build it from animated SVG paths that draw themselves plus a marker hand that tracks the drawing tip, then sequence elements so the board fills up in narration order.

When to use

  • A "scribe" / doodle explainer where a hand draws art and text onto a whiteboard.
  • Animating a sketch or logo being drawn live, line by line.
  • Handwriting that writes itself in, synced to a voiceover.
  • Turning a static illustration into a guided, stroke-by-stroke reveal.

Core mechanic: SVG stroke draw-on

A path "draws itself" by animating stroke-dashoffset from its full length down to 0 — the dash gap shrinks, exposing the stroke from start to end.

js
const path = document.querySelector("#stroke");
const len  = path.getTotalLength();
path.style.strokeDasharray  = len;       // one dash as long as the whole path
path.style.strokeDashoffset = len;       // pushed fully out of view = invisible
path.getBoundingClientRect();            // force reflow before transitioning
path.style.transition       = "stroke-dashoffset 1.2s linear";
path.style.strokeDashoffset = "0";       // draws start → end

Use pathLength="1" on the <path> to normalize: then stroke-dasharray:1; stroke-dashoffset:1 draws any path regardless of real length, and a single eased value (0→1) is the draw progress. linear (or a very gentle ease) reads as a steady hand; avoid ease-in-out on long strokes — it makes the marker lurch.

The marker hand follows the tip

The illusion lives or dies on the hand sitting at the exact point being drawn. Sample the path at the current draw progress and place the hand image there each frame; the pen-nib of the hand asset must align to the path point (offset the image so its tip, not its corner, lands on the point).

js
const hand = document.querySelector("#hand");   // <img>/<g>, tip at its top-left "nib"
const NIB = { x: 6, y: 4 };                      // nib offset inside the hand art
function drawOn(path, hand, dur = 1200) {
  const len = path.getTotalLength(), t0 = performance.now();
  path.style.strokeDasharray = len;
  (function frame(now) {
    const k = Math.min(1, (now - t0) / dur);     // 0→1 progress
    path.style.strokeDashoffset = len * (1 - k); // expose start→tip
    const p = path.getPointAtLength(len * k);    // current tip in SVG coords
    hand.setAttribute("transform", `translate(${p.x - NIB.x} ${p.y - NIB.y})`);
    if (k < 1) requestAnimationFrame(frame);
  })(t0);
}

getPointAtLength(len * k) is the workhorse: it returns the on-path coordinate at the same progress driving strokeDashoffset, so the nib stays glued to the growing stroke. Lift the hand (fade out, jump to the next start) between separate strokes so it doesn't slide across blank board.

Reveal order — the board fills up

Sequence elements the way a person would draw the scene, in narration order: each illustration and each handwriting block draws in, holds, then the next begins. Stagger strokes within one drawing; hold a beat after each element lands before moving the hand on. One drawing = one beat of the VO. Don't draw two things at once — the eye (and the single hand) can only follow one tip.

StepMechanism
Draw a strokestroke-dashoffset len→0, hand at getPointAtLength(len*k)
Multi-stroke drawingstagger strokes; hand hops to each new stroke's start
Handwritingper-letter/word strokes drawn in reading order, hand following
Holdpause hand off-board after element completes (0.4–1.0s)
Next elementmove/fade hand to next start; begin its draw-on
Erase / clearwipe or reverse-draw to clear the board for the next section

Converting art to drawable paths

Draw-on needs single-stroke, open paths in draw order — not filled compound shapes.

  • Author or trace line art as open strokes (Illustrator/Inkscape: outline, not fill). A filled blob has no "stroke" to animate; convert fills to a centerline stroke or draw an outline path and fill it after the outline completes.
  • Order matters: the <path> elements should appear in the SVG in the order a hand would draw them — animate them in document order.
  • Long continuous lines look most natural; break only where a real pen would lift.
  • For solid color regions: draw the outline stroke first, then fade/grow the fill in behind it (opacity or a clip-reveal) so it reads as "colored in after."
  • Handwriting: use a stroke/handwriting font, convert glyphs to paths, then draw each glyph's path in stroke order. Or fake it: mask the word and sweep a reveal left-to-right with the hand riding the mask edge.

Marker hand asset

A PNG/SVG of a hand holding a marker, with the nib at a known point. Measure the nib offset once (NIB above) and reuse it. Keep one hand asset for the whole piece for consistency. Mirror it (scaleX(-1)) only if you need a left hand. Add a faint shadow under the hand for depth; drop it during holds when the hand is "off."

Erase / wipe transitions

Between sections, clear the board so it doesn't get cluttered:

  • Wipe: animate a clip-path (or a white rectangle) across the filled group to wipe it away, hand optionally "erasing."
  • Reverse draw-on: run strokeDashoffset 0→len to un-draw, hand following backward.
  • Fade + redraw: cross-fade the old group out and start the next group's draw-on.

Keep one clear grammar (always wipe left-to-right, say) so transitions feel deliberate.

Pacing to narration

Draw-on duration should track the spoken line, not a fixed timer. Estimate from VO at ~2.3 words/sec; set each drawing's total draw time to roughly the length of the line it illustrates, plus a short hold. A label that the VO names should finish drawing right as the word is spoken. Slow the hand for emphasis lines; speed minor connective strokes. Never let the hand idle on a finished board while narration continues — move it off or start the next element.

Output checklist

  • Every stroke is a single open path in draw order; nothing pops in fully formed.
  • The hand nib sits exactly on the draw tip throughout each stroke (no float, no lag).
  • Hand lifts cleanly between separate strokes/elements; no sliding across blank board.
  • Elements draw in narration order, one at a time; each holds before the next.
  • Handwriting draws in reading order; fills appear after their outline.
  • Erase/wipe grammar is consistent; the board never over-clutters.
  • Draw timing tracks the VO line it illustrates.
  • prefers-reduced-motion: show the completed board (all strokes at offset 0, hand hidden) without the drawing animation.

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 self-contained whiteboard scene (draw-on illustration, handwriting, a short scribe beat) the deliverable is one HTML file that opens directly in a browser — no build step, no render pipeline. A single file is the right tier for web motion; don't reach for a bundler when one file does the job. (For a full narrated scribe video with baked VO, build it as a Remotion composition and verify via remotion still — see explainer-video.)

Output contract:

  • One .html file: SVG inline (paths in draw order, pathLength="1" or measured length), the hand asset inline or as a data-URI, and the draw-on driver in one inline <script>.
  • All drawing on one master timeline (GSAP tl, or one rAF clock) so the whole sequence has a single playhead you can seek — strokeDashoffset AND the hand position both driven by the same progress value.
  • Include the seek harness below so any moment can be frozen for screenshots. Freezing at t shows exactly how much of the path is drawn — the verification leans on this.

Seek harness — freeze an exact moment for screenshots. ?t=N seeks the master timeline to N seconds and pauses, so a screenshot lands on a still, deterministic frame mid-draw.

html
<script>
  // ... build your master timeline as `tl` (drawOn calls placed on it) ...
  const t = new URLSearchParams(location.search).get("t");
  if (t !== null) { tl.pause(); tl.seek(parseFloat(t)); }  // frozen at t seconds
  // no ?t → plays normally
  window.__ready = true;            // ready signal for headless wait
  console.log("duration", tl.duration());
</script>

Verify loop — render → freeze → screenshot → check:

  1. Open the file at moments across the timeline — start, a frame mid-draw, end: …/scribe.html?t=0, ?t=<dur/2>, ?t=<dur>. Read tl.duration() from the console for the end time. Pick at least one t in the middle of a stroke — that is where the hand-on-tip illusion is proven or broken.
  2. Headless-screenshot each frozen frame:
    bash
    npx playwright screenshot --wait-for-timeout=500 "file://$PWD/scribe.html?t=1.2" frame-mid.png
  3. INSPECT each still — check fidelity: at a mid-stroke t the hand nib sits exactly on the leading tip of the drawn portion (not ahead, not behind, not floating); the path is drawn only up to the tip; at ?t=<dur> every stroke completes cleanly with no gaps, no overshoot, no leftover dash. Check artifacts: hand sliding across blank board between strokes, nib offset wrong (tip floats off the line), strokes drawing out of order, fills appearing before their outline, clipped/off-canvas drawings, FOUC before fonts load.
  4. Iterate: adjust NIB, stroke order, durations, and re-screenshot until each frozen frame looks like a hand mid-draw.

Before you finish:

  1. Opens standalone in a browser — no console errors, no missing CDN/assets.
  2. One master timeline; ?t=N freezes correctly and strokeDashoffset + hand position share one progress value.
  3. Screenshotted at start / mid-stroke / end — hand nib on the tip throughout, strokes complete cleanly at the end, draw order correct.
  4. Hand lifts between strokes; no sliding across blank board.
  5. prefers-reduced-motion shows the completed board (all strokes drawn, hand hidden) without the drawing animation.

For a narrated, frame-deterministic export, port the same draw-on (drive strokeDashoffset and the hand transform from useCurrentFrame()/fps) into a Remotion composition and render to MP4/GIF — see the explainer-video skill for the render-stills → encode loop.

Reference files

  • references/draw-on-recipes.md — runnable draw-on mechanics: getTotalLength/getPointAtLength hand-follows-path, GSAP DrawSVGPlugin and MotionPathPlugin variants, multi-stroke staggering, per-glyph handwriting, fill-after-outline, erase/wipe and reverse-draw transitions, and the Remotion frame-driven port.
  • references/whiteboard-pipeline.md — turning art into drawable single-stroke SVGs, draw-order authoring, handwriting/stroke fonts, the marker hand asset and nib calibration, pacing draw time to the VO line, and the storyboard-to-board build sequence.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Whiteboard Animation AI skill do?

This skill should be used when the user asks to "make a whiteboard animation", "draw-on explainer", "VideoScribe style", "hand-drawn explainer video", "animate a sketch being drawn", "do a doodle/scribe video", "show a hand drawing the illustration", or "animate handwriting appearing on a whiteboard". Covers SVG stroke draw-on, a marker hand that follows the path tip, staggered reveal, erase/wipe transitions, and narrated export.

Why use Whiteboard Animation on TypingMind?

Because you install it once and use it with any model. Whiteboard Animation is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Whiteboard Animation in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/iart-ai/explainer-video-skills/tree/main/skills/whiteboard-animation. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Whiteboard Animation?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Whiteboard Animation?

As many as you like. As long as a model supports skills, you can use Whiteboard Animation with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Whiteboard Animation AI skill free?

Yes. It is published on GitHub by iart-ai under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇