About this skill
A marble rolls down a ramp and starts a curved chain, which feeds a fan that widens into a grid. The back faces of the dominoes are coloured, so the fallen field spells a word, a logo or any picture you draw as a text mask. A layout script places every domino from a word or a mask, Blender's rigid bodies bake the fall, and a report names the first domino that stayed standing with the change to try. The agent makes that change and runs the bake once more. A second script films the result with a camera that follows the dominoes that actually fell.
You say
Using this skill, build a domino run in Blender that spells our brand name when it falls. Show me the domino count first, then bake it, fix any stalls, and render a 12-second clip at 1280x720.
How it works
- Tell your agent the word or logo, the colours, and whether you want a still, a clip or both.
- It draws the picture as a text mask if the built-in font can't, and shows you how many dominoes that takes.
- It builds the scene in Blender, bakes the fall and reads the report. If a domino stayed up, it closes the gaps near it and runs the bake once more.
- It previews the camera on a few frames, renders with Eevee and encodes the clip with ffmpeg.
Getting a good result
- Keep the picture narrow. The fan grows with the square of the width, so a short word bakes far faster than a long one.
- Bake a two-letter test first. It takes seconds, so it is the cheap place to find a bad setting.
- Use two strongly contrasting colours. Fallen dominoes lean on each other, so each pixel is a tilted face, not a flat tile.
What you get
- A layout generator for the chain, fan and picture, from a word or a text mask
- A Blender script that builds the scene, adds the rigid bodies, bakes and reports where the wave stopped
- A film script with a camera that follows the fall and rises over the picture
- Stability settings with the reason for each, and a retake table of what you see and what to change
Good fit
- Logo and name reveals for launches, intros and social clips
- Learning Blender rigid-body physics from a working scene
- Agents that drive Blender headless or through an MCP connection
Not for
- Real-world domino builds: the sizes are tuned for the simulation
- Photoreal product shots, which need modelling and lighting work beyond this scene
- Large photos as pictures: each pixel is a domino, so detail stays coarse
How to install it
- Download the ZIP and unzip it.
- Move the folder holding
SKILL.mdinto your agent's skills folder, such as~/.claude/skills/for Claude Code. - Start a new session and ask for what you want to make.
Needs: An agent that reads skills, Blender 5.x, tested on 5.2.3, ffmpeg, to turn the frames into a video.
Popular on X
Posts showing demand for this kind of video. Unless a post names this tool, it wasn't made with it.
- A Blender domino field built and baked from one message 1,912 views on X, measured 7 November 2026
FAQ
What does it cost?
$2.99 once for this skill, or $99 once for every skill in the library. It also comes with every paid AdAnt plan, from $15 a month. It runs on your own computer and uses no AdAnt credits. It runs inside Blender, which you install separately.
Can I use it in commercial work?
Yes. You can use the skill and what it makes in your own commercial work, including for clients. You can't resell the skill.
How long does a bake take?
The field in the clip has 948 dominoes and baked 360 frames in about 10 seconds on an Apple M-series laptop, and Eevee rendered its 319 frames at 1280x720 in 72 seconds. Bake time grows with the number of dominoes, and the fan that widens the wave grows with the square of the picture's width.
What if some dominoes stay standing?
The fall report names the first domino that stayed standing and suggests a change, usually tighter spacing in the rows around it. The agent bakes again with that one change, and stops after three rounds to show you the report. In testing, a fan with too wide a side gap went from 118 to 309 of 309 dominoes down in two rounds.
Can my agent use it through Blender MCP?
Yes, by calling the same function the command line uses inside a running Blender. That path is untested, and it resets Blender to an empty file first, so the skill tells the agent to save your open file or ask before running it. The headless path with blender -b is the tested one.
Did the author of the domino post write this skill?
No. The idea comes from a post linked on this page, in which Blender built and baked a domino field from one message. Its prompt states no license, so AdAnt wrote this skill, its scripts and its picture, and the author did not write or review it. The clip on this page was made by AdAnt with this skill.
Can I use what it makes commercially?
Yes. What you make, such as .blend files, stills and videos, is yours. The two scripts that use Blender's Python API are GPL-3.0-or-later, as Blender asks of published scripts that use it, and the layout script is MIT. Both license texts are in the ZIP.
Related skills

Jolt Physics Simulation
Simulate thousands of rigid bodies with Jolt and render the collapse, repeatably, frame by frame.

Blender Scenes through MCP
Have your agent build a Blender scene one step at a time, look at each result, and render the finished shot.

Physics Simulation One-Take Film
Simulate lightning, a crack or a fire front, then ride it in one unbroken camera shot, with sound from the simulation and a flash safety check.

Visible Music Machine
Write a tune as notes first, then have your agent build a machine that plays it, where every note is a ball or mallet hit you can see, with sound on the exact frame.

Shader Reveal Site
Have your agent build a page where the pointer or the scroll uncovers a second picture through a shader, then measure the trail, the GPU and every fallback in Chrome.

Procedural Character Performance
Animate an original character dancing, walking or trotting to a beat entirely in code, with feet that never slide and every hit on the beat.

Cursor Playground Hero
Have your agent build a hero section visitors play with, on small simulations that behave the same on every screen and never blow up.

Logo Reveal and Ident
Have your agent turn your SVG logo into a 3 to 8 second reveal with a sound hit, as wide, vertical, square and transparent videos that end on your logo, checked against the SVG.

Reverse Any Video: Rebuild a Video as an Editable HyperFrames Project
Have your agent rebuild a video file as an editable HyperFrames project, then check its render against the original frame by frame.

Code-Rendered 3D Animated Short
Have your agent write a tiny wordless story, build its characters and set in three.js, shoot it from a shot list, score it and render a frame-exact MP4.

Hand-Painted Brushstroke 3D
Have your agent repaint a live three.js scene as brush strokes that stay on their surfaces when the camera moves, then measure it and record it.
Pixel-Art Sprite Film
Have your agent make a 15 to 40 second pixel-art scene in code, with sprites, parallax and a chiptune score, rendered to an MP4 whose pixels stay crisp.