explaingit

opencoworkai/open-codesign

5,800TypeScriptAudience · vibe coderComplexity · 2/5LicenseSetup · easy

TLDR

A free, open-source Electron desktop app that turns text prompts into visual designs, HTML, PDF, PowerPoint, using your own AI API keys, running locally with no cloud workspace.

Mindmap

mindmap
  root((open-codesign))
    What it does
      Text to design
      Live preview
      Export files
    Features
      Iterative edits
      Comment pins
      Version history
    Tech
      Electron app
      TypeScript
      20 AI providers
    Use cases
      UI prototypes
      Slide decks
      HTML exports
Click or tap to explore — scroll the page freely

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Things people build with this

USE CASE 1

Generate a slide deck from a text description and export it as a PowerPoint file in minutes.

USE CASE 2

Iterate on a UI design by clicking individual elements and describing changes without rewriting the whole prompt.

USE CASE 3

Run AI design generation offline using Ollama models so no data leaves your machine.

USE CASE 4

Create an HTML prototype from a plain-English brief and export it as a ZIP archive for a developer to implement.

Tech stack

TypeScriptElectron

Getting it running

Difficulty · easy Time to first run · 30min

Requires an API key from at least one AI provider such as Claude, OpenAI, or Gemini, or a local Ollama installation.

Use freely for any purpose, including commercial use, as long as you keep the copyright notice.

In plain English

Open CoDesign is a free, open-source desktop application that lets you turn text prompts into visual designs, prototypes, and documents using AI. You describe what you want in plain English, the app sends that to an AI model of your choice, and the result appears as a live preview. The output can be exported as HTML, PDF, a PowerPoint file, a ZIP archive, or Markdown. The app is positioned as a local-first alternative to cloud-based AI design tools. It runs as an Electron desktop app on your own computer, meaning your data does not pass through a cloud workspace by default. You bring your own API keys for whichever AI provider you use: Claude, GPT-4, Gemini, DeepSeek, Ollama (for locally-run models), or any OpenAI-compatible endpoint. You can also sign in directly with a ChatGPT Plus or Pro subscription to use Codex models without entering a key. Beyond the initial generation step, the app supports iterative refinement. You can click on any element in the preview to drop a comment pin on it, then describe the change you want, and the model rewrites only that region. The app also surfaces a panel of parameters the AI identifies as worth adjusting, such as colors or spacing, so you can tune those without writing another full prompt. Each session is saved locally with a version history. The README compares Open CoDesign directly to Claude Design, v0 by Vercel, Lovable, and Bolt.new, highlighting that it is free to download, runs offline, supports more than 20 model providers, and exports editable files. The project is MIT-licensed and available for Mac, Windows, and Linux.

Copy-paste prompts

Prompt 1
Using Open CoDesign with my Claude API key, generate a landing page for a SaaS productivity tool aimed at small teams, export the result as HTML.
Prompt 2
I have a generated design in Open CoDesign, show me step by step how to use the comment pin feature to refine only the hero section.
Prompt 3
Set up Open CoDesign with Ollama so I can generate designs locally without sending any data to a cloud provider.
Prompt 4
Generate a pitch deck for a mobile recipe app in Open CoDesign and export it as a PowerPoint file ready for investors.
Prompt 5
I have both Open CoDesign and v0 by Vercel available, when should I use each one, and what does Open CoDesign do that v0 can't?
Open on GitHub → Explain another repo

← opencoworkai on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.