A Visual Design that ships with your code. Agents and humans work together on the same canvas, where every change is a diffable JSX stored in a .uidx file inside your repo. Built on OpenPencil and Ski
Design in your repo. A visual design workspace for your project.Editable on a canvas. Readable by people and agents. Versioned with your code.
Workflow illustration: one design, shared by the canvas, your files, and your tools.
UIDX brings a design canvas into your application's repository. Install @uidxkit/uidx as a local development dependency, run npm run uidx, and work on designs stored in your project's .uidx/ folder.
A .uidx file combines Markdown describing a design's purpose with a declarative node tree describing its appearance. Edit the canvas and the file updates. Edit the file and the canvas follows. Changes produce focused text diffs you can review in a pull request.
- A visual workspace. Create pages, draw shapes, edit text, and adjust layout and styling from the canvas and inspector. - Reusable design systems. Components, instances, variants, slots, and design tokens share a document namespace. - Your project owns the files. Keep designs, assets, and fonts next to your application and version them with Git. - A shared workflow for people and agents. The CLI and MCP bridge let external tools inspect and edit the same files the viewer uses. - Local by default. The viewer and project server run on your machine. Canvas synchronization does not require an LLM or a hosted design service.
UIDX is an early release. The file format and APIs may change before 1.0. It is a design workspace; it does not render your application's existing UI components as Storybook stories.
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/uidxkit--uidx.html)
Open-source control plane for Codex projects: Git-backed context, visible agent progress, scoped MCP access, resumable work, and safe handoffs.
NiazMorshed2007/jev-reviewLocal-first MCP plugin for continuous software-quality review by AI coding agents, powered by Jev.
nassim-arifette/jevgrepJev-powered semantic code search for coding agents — find behavior across repositories via CLI or MCP, with exact source excerpts and line numbers.
LilithGames/a2waveTurn the agent CLIs you already use into shared services your whole team can call. Natural-language-driven agent orchestration for Claude Code, Cursor Agent, and OpenAI Codex.
pranee54/AgentDoctorLocal CLI that audits coding-agent configuration for security, instructions, context, and MCP — no API key or code upload by default.
dziksu/StructSmithLocal-first, MCP-native tool for modelling software architecture. C4-style semantic model, React Flow editor, SQLite, single container — a self-hosted Structurizr alternative your AI client can edit o
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.