denysosadchyi

denysosadchyi / figmosha

Public

AI-driven Figma design automation — Claude Code draws UI via Playwright + Scripter plugin

11
0
100% credibility
Found Apr 08, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Figmosha automates UI component creation in Figma by bridging AI-generated scripts to a browser-controlled Figma plugin.

How It Works

1
🔍 Discover Figmosha

You hear about a handy tool that lets AI automatically draw and build user interface designs right inside your Figma files.

2
🛠️ Prepare your Figma workspace

You add a simple helper extension inside Figma to make it ready for the magic.

3
🚀 Start the design helper

You launch a small program on your computer, sharing your Figma file link and login details so it can connect everything.

4
🤖 Team up with AI

You chat with an AI like Claude, describing the button or layout you want, and it sends perfect drawing instructions to Figma.

5
✨ Watch your designs come alive

In real time, frames, texts, colors, and layouts appear on your Figma canvas just as you imagined.

6
🔄 Tweak and build more

You send more ideas or fixes, and everything updates smoothly, even running other helpers for properties or comments.

🎉 Pro designs ready to use

Your Figma file is now packed with polished UI components, saving you hours of manual work.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 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 figmosha?

Figmosha is an AI-driven Figma automation tool built in Python that uses Playwright to control a browser, log into Figma, and execute scripts via the Scripter plugin. You start a server with a Figma file URL, credentials, and your OS display, then pipe in Figma API code from the CLI—like `python run.py "figma.createRectangle()"`—to draw UI components, frames, text, and auto layouts directly on the canvas. It tackles manual design tedium in AI-driven development by letting Claude generate spec-driven code that builds production-ready designs with variables, text styles, and Propstar grids.

Why is it gaining traction?

It stands out with pipelines for adding components, importing PDF slides as frames, and resolving Figma comments via REST API—all verifiable via text dumps, skipping slow screenshots. Devs dig the atomic approach: build structures first, bind variables second, ensuring reliable AI-driven Figma scripting without silent failures. CLI hooks for plugins like Propstar and reopening Scripter make it a seamless bridge for Claude code to live designs.

Who should use this?

Frontend devs prototyping UIs from AI specs, design engineers automating component libraries, or solo makers converting code sketches to Figma files. Ideal for teams in AI-driven dev GitHub workflows needing quick spec-driven AI design iteration without hand-drawing every button.

Verdict

At 11 stars and 1.0% credibility, Figmosha feels experimental—docs are solid with pipelines but lack tests and broad Linux fixes. Try it for AI-driven Figma hacks if you're comfy with Playwright setup; otherwise, wait for maturity.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.