Hand-drawn explainers, without the hand cramp
The whiteboard genre, rebuilt for agents: stick-figure rigs, pose tracks, draw-on reveals and captions in one editable film file — with five drawn looks on the same timeline.
made with coanimator
Drawn films from the template
Rendered by the app from the drawing-animation sample projects — pose tracks, narration and captions in one editable film file.
how it's built
How to make one in CoAnimator
- 01
Describe it
A brief in plain language: what the video shows, the moments that matter, the tone. No timeline surgery.
- 02
The agent builds the scenes
Claude Code, Codex, Gemini or ours — the agent authors each scene as editable code on the templates and skills the studio ships.
- 03
Narration lands timed
The built-in voiceover plugin renders the narration part by part and places every line on the timeline where it belongs.
- 04
Review in the editor
Timeline, media panel, live stage. Steer with feedback — "slow the intro", "swap that screenshot" — and watch it hot-reload.
- 05
Render locally
MP4 exports run on your machine, in-app or headless via the coa CLI. Unlimited, watermark-free, nothing uploads anywhere.
Five looks, one film
Ink & pencil
Clean ink lines or sketchy pencil — the classic whiteboard read, driven by the same pose data.
Illustrated & watercolor
Filled illustration and two watercolor treatments when the explainer should feel warmer than a marker.
Draw-on reveals
Outlines that draw themselves around subjects and scenes — the signature whiteboard move, on real footage too.
Old tools vs. this
Legacy whiteboard software gives you a library of stock hands dragging clip-art. Here the agent authors the film — characters, poses, captions, camera — and switching the entire look is a dropdown, not a re-purchase. Finished sample films ship in the template. See the templates.
FAQ
What’s the best way to make a whiteboard animation video?
The classic routes are drawing it yourself, hiring an illustrator, or using legacy whiteboard software with stock clip-art hands. CoAnimator adds a fourth: an AI agent authors the film — characters, poses, draw-on strokes, captions, camera cuts — as one editable data file, in your choice of ink, pencil, illustrated, or watercolor looks. You describe the explainer, review it on a live stage, and render the MP4 on your own machine.
How does an AI agent make a hand-drawn explainer?
Strokes in the drawing-animation template draw themselves along their own path, the way a hand would ink them, and characters are stick-figure or illustrated rigs with pose tracks and lip sync. The agent directs all of it — scenes, poses, captions, timing — and every part stays editable. Change a pose or a line of narration and the film re-renders without redrawing anything.
Can I edit a whiteboard video after it’s made?
Yes — the film is data, not baked frames. Ask the agent for a new scene, a different pose, tightened pacing, or an entirely different look, and re-render locally; switching the whole film from ink to watercolor is a setting, not a re-purchase. That keeps training and onboarding explainers current instead of aging out with the first product change.
Does CoAnimator use stock clip-art hands like other whiteboard tools?
No — there’s no library of stock hands dragging clip-art. Draw-on reveals are real strokes that trace themselves around subjects and scenes, and they work over real footage too, so you can sketch an outline around a live product shot. The signature whiteboard read is there; the canned look isn’t.
How is CoAnimator different from traditional whiteboard animation tools?
Traditional whiteboard tools are asset assemblers: you pick from their image library and the software drags a stock hand across the canvas. In CoAnimator an agent authors an original film from your script, with characters and scenes stored as data you can edit. The trade is honest — a stock-asset tool is quicker for a one-off list video, while an authored film is stronger when the explainer has to carry your story and survive revisions.
How does narration stay in sync?
The standard voice loop: TTS parts rendered per beat, placed on the timeline, scenes retimed to the real audio.