onlook
The Cursor for Designers • An Open-Source AI-First Design tool • Visually build, style, and edit your React App with AI
Install / Use
npx skills add onlook-dev/onlookInstalls into whichever agent you are using.
.cursorrules
Cursor IDE rules (legacy)
Quality Score
Category
DesignSupported Platforms
Skill content
View source on GitHub中文 | Español | Deutsch | français | Português | Русский | 日本語 | 한국어
</div>This is the open-source visual editor that started Onlook (Next.js + Tailwind). The hosted product we're building next, for AI-native designers, is in early access via the waitlist above.
An Open-Source, Visual-First Code Editor
Craft websites, prototypes, and designs with AI in Next.js + TailwindCSS. Make edits directly in the browser DOM with a visual editor. Design in realtime with code. An open-source alternative to Bolt.new, Lovable, V0, Replit Agent, Figma Make, Webflow, etc.
What you can do with Onlook:
- [x] Create Next.js app in seconds
- [x] Start from text or image
- [x] Use prebuilt templates
- [x] Visually edit your app
- [x] Use Figma-like UI
- [x] Preview your app in real-time
- [x] Manage brand assets and tokens
- [x] Create and navigate to Pages
- [x] Browse layers
- [x] Manage project Images
- [x] Detect and use Components – Previously in Onlook Desktop
- [ ] Drag-and-drop Components Panel
- [x] Use Branching to experiment with designs
- [x] Development Tools
- [x] Real-time code editor
- [x] Save and restore from checkpoints
- [x] Run commands via CLI
- [x] Connect with app marketplace
- [x] Deploy your app in seconds
- [x] Generate sharable links
- [x] Link your custom domain
- [ ] Collaborate with your team
- [x] Real-time editing
- [ ] Leave comments
- [ ] Advanced AI capabilities
- [x] Queue multiple messages at once
- [ ] Use Images as references and as assets in a project
- [ ] Setup and use MCPs in projects
- [ ] Allow Onlook to use itself as a toolcall for branch creation and iteration
- [ ] Advanced project support
- [ ] Support non-NextJS projects
- [ ] Support non-Tailwind projects
Getting Started
Use our hosted app or run locally.
Usage
Onlook will run on any Next.js + TailwindCSS project, import your project into Onlook or start from scratch within the editor.
Use the AI chat to create or edit a project you're working on. At any time, you can always right-click an element to open up the exact location of the element in code.
<img width="600" alt="image" src="https://github.com/user-attachments/assets/4ad9f411-b172-4430-81ef-650f4f314666" /> <br>Draw-in new divs and re-arrange them within their parent containers by dragging-and-dropping.
<img width="600" alt="image" src="assets/insert-div.png"> <br>Preview the code side-by-side with your site design.
<img width="600" alt="image" src="assets/code-connect.png"> <br>Use Onlook's editor toolbar to adjust Tailwind styles, directly manipulate objects, and experiment with layouts.
<img width="600" alt="image" src="assets/text-styling.png" />Documentation
For full documentation, visit docs.onlook.com
To see how to Contribute, visit Contributing to Onlook in our docs.
How it works
<img width="676" alt="architecture" src="assets/architecture.png">- When you create an app, we load the code into a web container
- The container runs and serves the code
- Our editor receives the preview link and displays it in an iFrame
- Our editor reads and indexes the code from the container
- We instrument the code in order to map elements to their place in code
- When the element is edited, we edit the element in our iFrame, then in code
- Our AI chat also has code access and tools to understand and edit the code
This architecture can theoretically scale to any language or framework that displays DOM elements declaratively (e.g. jsx/tsx/html). We are focused on making it work well with Next.js and TailwindCSS for now.
For a full walkthrough, check out our Architecture Docs.
Our Tech Stack
Front-end
- Next.js - Full stack
- TailwindCSS - Styling
- tRPC - Server interface
Database
AI
- AI SDK - LLM client
- OpenRouter - LLM model provider
- Morph Fast Apply - Fast apply model provider
- Relace - Fast apply model provider
Sandbox and hosting
- CodeSandboxSDK - Dev sandbox
- Freestyle - Hosting
Runtime
Contributing
If you have a suggestion that would make this better, please fork the repo and create a pull request. You can also open issues.
See the CONTRIBUTING.md for instructions and code of conduct.
Contributors
<a href="https://github.com/onlook-dev/onlook/graphs/contributors"> <img src="https://contrib.rocks/image?repo=onlook-dev/onlook" /> </a>Contact
- Team: Discord - Twitter - LinkedIn - Email - YouTube
- Project: https://github.com/onlook-dev/onlook
- Website: https://onlook.com
License
Distributed under the Apache 2.0 License. See LICENSE.md for more information.
<!-- https://www.markdownguide.org/basic-syntax/#reference-style-links -->Related Skills
headroom
73.4kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
ruflo
73.0k🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
guizang-ppt-skill
26.7kAI-agent Skill for generating polished HTML slide decks: editorial magazine and Swiss layouts, image prompts, social covers, and a WebGL/low-power presentation runtime.
Memori
16.9kMemori is agent-native memory infrastructure. A LLM-agnostic layer that turns agent execution and conversation into structured, persistent state for production systems.
