Design 3D-printable parts by talking

> "It asks for the real dimension instead of inventing one."

Shortly after

> "Make an adapter that connects my shop vac hose to the dust port on my table saw" > "Done: Ø57.6 → Ø35.0 · 74 mm"

is interesting. When did it ask for the dimensions?

ur-whale an hour ago | parent | prev | next [-]

> runs on your computer

Only if you have a cupertino-made toaster that is.

noduerme an hour ago | parent | prev | next [-]

which part of this is the innovation?

protocolture an hour ago | parent | prev | next [-]

Looks cool.

I have had issues with language CAD stuff before, often not performing as advertised and would love to tinker with it.

And whats great here is that I cant test the web version, and because its for mac silicon only, I also cant test the local version.

So Kudos? I guess?

	▲	okramcivokram 43 minutes ago \| parent [-]
		I had some good results using CadQuery [1] with CQ-editor [2] for visualization driven by Claude Code enhanced by a cadquery-llm-skill [3]. I didn't make anything too complicated, just cases for various ESP32 projects I am tinkering with, but that was very easy just describing what I want and iterating and refining until I am satisfied. It was much easier than doing it manually in FreeCAD and more dynamic as the changes were cheap and not tedious to do. [1] https://github.com/cadquery/cadquery [2] https://github.com/CadQuery/CQ-editor [3] https://github.com/jmwright/cadquery-llm-skill

> You talk. It models. Physics checks the work.

How does physics "check the work"? Do you run a physics engine? Do you make it in reality and do physical tests? What the fuck?

Gets this slop away from me.