ganjihong cf9cd0366a refactor(cdsl_engine): extract topology evidence export from the adapter
Phase 6 of the decoupling refactor (behavior-preserving move):
- topology_export.py: body_geometry / surface_geometry / topology_records
  (~290 lines) moved verbatim out of build123d_adapter.py as module-level
  functions; includes the circle_center_mm/radius_mm concentric-circle
  evidence added by the lk_dev integration
- build123d_adapter.py: the three methods are now one-line forwards, so
  kernel construction and topology evidence export live in separate files
  and can evolve independently

No protocol change (GeometryAdapter untouched). Verified by:
- golden comparison: full-corpus analyze (2881 docs) + 12 complete
  rebuilds -- zero field differences vs f79c287
- full test suite: 435 tests, failure set identical to baseline
2026-09-09 16:07:25 +08:00
2026-08-24 14:52:08 +08:00
2026-09-02 13:51:35 +08:00
2026-09-04 11:17:36 +08:00
2026-08-19 19:34:30 +08:00
2026-08-21 14:12:27 +08:00
2026-08-21 14:12:27 +08:00
2026-08-21 14:12:27 +08:00

CAD CDSL Workspace

This repository is organized as a CAD generation workspace with four runtime programs and one AI skill:

  • frontend/: Agent UI and 3D model preview.
  • backend/: API, generation orchestration, engine, and official CDSL library.
  • solidworks_to_json/: SolidWorks export plugin.
  • json_to_cdsl/: SolidWorks JSON to parameterized CDSL converter.
  • backend/agent/skills/cad-engine/: Instructions for the backend AI agent to use the CAD engine.
  • cdsl 5/: Existing experimental assets kept unchanged for reference.

Run both application services from the repository root with:

./run.sh

On Windows, run run.bat instead.

Python 3 and Node.js (including npm) must be installed first. Both startup scripts create backend/.venv and install the backend and frontend dependencies automatically when they are missing.

S
Description
No description provided
Readme 221 MiB
Languages
Python 67.2%
JavaScript 30.1%
TypeScript 2%
CSS 0.5%
Shell 0.1%