An AI agent skill that turns any subject into a 3D film or app
README: hero collage, stills for three more looks, style reference section
Nguyen Xuan An3d44bbfc533e
diff --git a/README.md b/README.mdindex 6441854..22be210 100644--- a/README.md+++ b/README.md@@ -1,9 +1,14 @@# viz3d+[](https://github.com/codepawl/viz3d/releases)+[](LICENSE)+Visualize anything in 3D. Give your agent a subject: a place, a system, an object, a product, a creature, a story or anidea. The agent researches it, builds models to real dimensions, and makes a film, an app you click through, a lesson,a book or a game. It checks its own frames before it shows you anything.++<table><tr><td width="200">@@ -120,11 +125,13 @@ default look for the mode.| | Dither + ASCII showcase | `ASCII` | A film that shows off a model, a tool or a product: the model stays clean, the world around it turns to ASCII || | Gradient glow ASCII | `glow` | A creature or object that is itself the art, drawn in glowing ASCII |-| Layer build | ASCII |+| Toy isometric | Illustrated cutaway plate ||---|---|+|  |  |+| **Point-cloud survey** | **Space-time cube** |+|  |  |+| **Layer build** | **ASCII** ||  |  |-| **Space-time cube** | |-|  | |Default looks: apps, prototypes, embeds and lessons use clean ops world. Films use dark documentary, showcases useDither + ASCII and books use the cutaway plate. You can show the agent a reference video of a new look. It studies the@@ -149,6 +156,22 @@ measures every gap before it records.The skill asks before it downloads anything, such as elevation data, map data or reference images.+### It learns the style from a picture++The agent never builds a look from words alone. It finds a reference in the same look, writes a style card (camera,+palette, light, brightness, density, hero props) and checks every round against it with measured gates:++- `compare-look.py` compares brightness, saturation and hue with the reference and fails when the render drifts.+- `ground-share.py` fails a frame where one flat colour covers more than a quarter of it.+- The subject must fill at least a third of the default view.+- Grass, trees and props come from CC0 and CC BY packs, never from code-built blades or boxes.++Every correction you give becomes a dated rule in a lessons file, so the next build starts from it.++++*Left: the first build. Right: the same subject after the agent applied its own review notes.*+## InstallIn Claude Code, add this repository as a plugin marketplace and install the plugin:@@ -258,6 +281,10 @@ The last command adds the 67js skills (`build-67js-scene` and `render-67js-media## What is new+- Style from a reference: a style card in the plan file, checked by measured gates (`compare-look.py`,+ `ground-share.py`).+- Ground cover from real assets: a CC0 ground texture under instanced grass tufts.+- Machines in cutaway plates move at their real gear ratios on one clock, in round cases with real plate finishes.- viz3d now covers any subject, not only things that are hard to understand.- The `showcase` mode: a film or stills that show off one object, product or model.- `/viz3d` with nothing after it shows the menu. The menu groups the modes by what people do with them and the looks by family.diff --git a/docs/images/canal.jpg b/docs/images/canal.jpgnew file mode 100644index 0000000..67ae14aBinary files /dev/null and b/docs/images/canal.jpg differdiff --git a/docs/images/hero.jpg b/docs/images/hero.jpgnew file mode 100644index 0000000..44b7933Binary files /dev/null and b/docs/images/hero.jpg differdiff --git a/docs/images/pointcloud.jpg b/docs/images/pointcloud.jpgnew file mode 100644index 0000000..a1e4ff6Binary files /dev/null and b/docs/images/pointcloud.jpg differdiff --git a/docs/images/self-review.jpg b/docs/images/self-review.jpgnew file mode 100644index 0000000..f438310Binary files /dev/null and b/docs/images/self-review.jpg differdiff --git a/docs/images/watch.jpg b/docs/images/watch.jpgnew file mode 100644index 0000000..cc7ee76Binary files /dev/null and b/docs/images/watch.jpg differ