HubLensLLMOpenCoworkAI/open-codesign
// archived 2026-05-02
OpenCoworkAI

open-codesign

AI🌱 NEW PROJECT BOOST#Electron#LLM#Design#Prototyping#Agentic
View on GitHub
34

// summary

Open CoDesign is an open-source, desktop-native application that allows users to transform prompts into polished prototypes and design artifacts locally. It supports a wide range of AI models through a bring-your-own-key approach, eliminating reliance on cloud-only subscriptions. The tool features interactive editing, responsive previews, and multi-format exports to streamline professional design workflows.

// technical analysis

Open CoDesign is an open-source, desktop-native AI design tool built on Electron that provides a local-first alternative to cloud-locked platforms like Claude Design or v0. Its design philosophy centers on user sovereignty, allowing developers to bring their own API keys for a wide range of LLMs while maintaining full control over their data and design artifacts. By offering a modular, agentic architecture that supports local file exports and iterative design refinement, it solves the problem of vendor lock-in and privacy concerns inherent in web-only AI prototyping tools.

// key highlights

01
Supports a wide range of LLM providers including Claude, GPT, Gemini, and local models via Ollama, ensuring no subscription lock-in.
02
Provides a local-first desktop experience that keeps all design state and artifacts on the user's machine rather than in the cloud.
03
Features a 'Comment Mode' that allows users to click specific UI elements to drop pins and request targeted AI rewrites.
04
Includes twelve built-in design skill modules that steer the model toward high-quality typography, spacing, and layout coherence.
05
Enables seamless one-click imports of existing configurations from tools like Claude Code or Codex for rapid onboarding.
06
Offers versatile export options including HTML, PDF, PPTX, ZIP, and Markdown to ensure generated prototypes are portable.

// use cases

01
Generating responsive web prototypes and UI components from text prompts
02
Refining design elements using AI-powered sliders and targeted region rewriting
03
Managing local design iterations with multi-model support and secure artifact exports

// getting started

To begin, download the appropriate installer for your OS (macOS, Windows, or Linux) from the GitHub releases page or use a package manager like Homebrew or winget. Upon first launch, navigate to the settings page to input your preferred LLM API key or configure a local Ollama endpoint. Once connected, you can select from fifteen built-in demos or enter your own prompt to start generating prototypes immediately.