The maker's introduction consists of the line above. Read the repository README for more details. Repository README ↗
Loading the full README.
Read the full README ↗Interactive polygon-modeling application. It executes immutable topology edits
through kami-engine-modeling and uploads arbitrary vertex/index meshes into
the real kotoba-lang/webgpu pipeline. The Pages build includes WebGPU-first /
WebGL2-fallback rendering, object and component selection, dimension-driven
sketch constraints, topology tools, BSP Boolean workflows, non-destructive
modifiers, UV/image materials, undo/redo, recovery and EDN/glTF exchange.
UI markup and styling are generated exclusively through kotoba-lang/html and
css.
The large-scene browser gates render 20,000 sphere instances / 11.2 million resident triangles as one shared-geometry draw. On the named local Chromium reference run, WebGPU measured 206.5 ms first submission and 0.2 ms p95 submit; forced WebGL2 measured 917.8 ms first submission and 6.6 ms p95 submit. Both report the same stable picking sample and source provenance. WebGL2 uses a cached mat4/color instance buffer rather than 20,000 individual draw calls.
Public app: https://kotoba-lang.github.io/kami-app-modeler/
The current 5/5 rating applies to the declared browser polygon-modeler scope.
The roadmap to commercial CAD/CAE maturity—NURBS/B-rep and STEP, assemblies,
associative manufacturing drawings, qualified CAE, a production modifier DAG,
very-large scenes and collaboration history—is governed by
kami-engine ADR-0049.
Verify the deployed application with:
MODELER_URL=https://kotoba-lang.github.io/kami-app-modeler/ npm run verify:webgpu
npm run verify:webgl2
Start with the user tutorial. Versioned production bundles
are created from v* tags only after the release verification gate succeeds.