Open-source Claude skill turns one image into a fully disassemblable 3D mechanical model
tinyfool · x · 2026-11-22
A demo circulating on X shows an open-source Claude skill that takes a single image and builds an intricately detailed, layer-by-layer disassemblable 3D mechanical model from scratch with Three.js.
- Using one ChatGPT-rendered image plus the single word "octopus," Claude produced a full mechanical design: a 10-blade mechanical iris with a 62mm ring inside the eye, a liftable 4mm borosilicate glass dome, and 8 mechanical arms with double rows of suction wheels extending into 16-segment spiral tips, all connected to a 48-tooth hub
- A teardown animation sheds one part every 3 seconds until the whole model fully disassembles at 19 seconds
- Notably, the octopus isn't among the skill's 3 preset animal templates, showing strong generalization
The quoted original article argues a workflow insight: instead of ordinary prompts, give Claude a "job description"-style structured brief — the difference is night and day.
More from coding & agent
- The 'Agent Platform of 2027' blueprint: actors, RBAC filesystems, inbox UX — yawnxyz · 2026-11-22
- Open-source tool unlocks third-party models in Claude Desktop via endpoint mapping — pneuny · 2026-11-22
- Salesforce's release-review agent cut review turnaround from 24 hours to 12 seconds — shoumikgoswami · 2026-11-22
- Claude can now design print-ready holiday cards — 7 prompts to replace Canva — aftahi_ai · 2026-11-22
- Dev workflow: architecture first, code last — why AI coding still eats your time — AryHHAry · 2026-11-22
- Neon's pleasant style can rub off on fable and other orchestrating models — olcan · 2026-11-22