ReshefElisha

ReshefElisha / jarvis-onshape-mcp

Public

Jarvis Onshape MCP — Claude Code plugin for driving real Onshape CAD. Vision-decomposition skill + truth-telling + parametric iteration + FeatureScript.

48
1
100% credibility
Found Apr 21, 2026 at 48 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A plugin for Claude that enables AI to create and edit 3D CAD models directly in Onshape with visual feedback and structured results.

How It Works

1
🔍 Discover your CAD helper

You find a friendly plugin that lets your AI assistant Claude create 3D designs for you in Onshape.

2
📱 Add it to Claude

With one simple command, you install the plugin into your Claude app so it's ready to use.

3
🔗 Link your design space

You connect your Onshape account once, securely, so Claude can build designs there.

4
💭 Describe your idea

You tell Claude in plain words what you want to make, like a phone stand or bracket.

5
🛠️ Watch it build

Claude draws sketches, adds shapes, shows pictures from every angle, and asks if it's right.

6
✅ Review and tweak

You check the renders and measurements, and Claude fixes anything off with your feedback.

🎉 Get your model

Your complete 3D design is ready to 3D print, share, or manufacture.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 48 to 48 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is jarvis-onshape-mcp?

This GitHub Jarvis project is a Python MCP plugin for Claude Code that lets AI drive real Onshape CAD: sketches, extrudes, fillets, mates, parametric iteration via variables, and custom FeatureScript. Drop a reference image or text prompt like "60x40x8mm plate with corner holes," and it builds, renders multi-view PNGs, extracts drawing dimensions via OCR, and returns truth on every mutation—status, changes, warnings with fix hints. Solves AI CAD's blind spots: no more silent failures or image misreads.

Why is it gaining traction?

Stands out with "truth-telling": real regen status, geometric diffs, interference checks—instead of faking success like most jarvis github ai wrappers. Vision-decomposition skill zooms/analyzes references first, parametric tools enable iteration (NEMA17 to NEMA23 swaps), and assembly mates plus FeatureScript escape hatch tackle pro designs. Renders and entity lists with outward normals make follow-ups reliable.

Who should use this?

CAD engineers prototyping mechanical parts in Onshape via natural language. AI devs building LLM agents for parametric modeling from drawings or photos. Teams automating iteration on assemblies with mates and custom features.

Verdict

Download from jarvis github for Onshape + Claude experiments—1.0% credibility score and 48 stars flag early maturity, but strong docs, install via `/plugin install github:ReshefElisha/jarvis-onshape-mcp`, and pytest coverage make it playable now. Solid copilot if you're the jarvis github llm guy; production needs more battle tests.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.