I made a Codex skill that turns a photo or an idea into a brick model

Hey, wanted to share a little project I've been working on. I had this idea a while ago: give an agent a photo or describe a building, and have it put together an actual 3D model out of individual bricks. With Astra I finally got somewhere with it. It's called klemm. You install the open-source skill in your own Codex setup and run it locally. It uses Blender and LDraw parts, and exports the model, rendered views, a PDF building guide and a parts list from the same model data. The cathedral on the site is one of my test models. Still early, and the examples haven't been physically build-tested yet. I'm going to keep working on the detail and output quality. I also got a bit carried away with the website 😅 That's what the video shows. Both the skill and site were built with Astra in Codex. klemm.build github.com/selamiselvi/klemmbausteine-skill This is also my submission for the Product Hunt Astra challenge, live today: producthunt.com/products/klemm