About Open Components
Open Components is a set of UI component guidelines covering user, developer, and agentic experience. Its MCP server gives agents tools to read the rules, contracts, and documentation section by section.
Key features
- 01Guidelines for UX, DX, and AX
- 02Component rules and contracts
- 03Documentation available as Markdown
- 04YAML contracts and JSON Schema
- 05Agent plugin with build and review skills
Use cases
- 01Check a component against accessibility and behavior rules
- 02Align APIs across a component library
- 03Give coding agents structured component requirements
- 04Validate custom contracts against the schema
MCP support
The MCP server exposes the component standard, rules, and documentation to agents as tools, with no sign-in required. An agent plugin can connect to the same server and adds skills for building and reviewing components.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.opencomponents.dev/mcpclaude mcp add --transport http com-uxfront-open-components https://mcp.opencomponents.dev/mcp{"mcpServers":{"com-uxfront-open-components":{"type":"http","url":"https://mcp.opencomponents.dev/mcp"}}}Open in Cursor{"servers":{"com-uxfront-open-components":{"type":"http","url":"https://mcp.opencomponents.dev/mcp"}}}Open in VS Code{"mcpServers":{"com-uxfront-open-components":{"type":"http","url":"https://mcp.opencomponents.dev/mcp"}}}{"mcpServers":{"com-uxfront-open-components":{"serverUrl":"https://mcp.opencomponents.dev/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
6| Parameter | Type | Required | Description |
|---|---|---|---|
| component | enum | Yes | The component or foundation, like button or design-tokens.design-tokens | button | spinner | input | card | checkbox | switch | textarea | tooltip | checkbox-group |
Details
From the project READMEOpen Components
Open Components is the standard for perfect UI components: guidelines for building UI components with a perfect user experience, developer experience and agentic experience, whether humans or AI agents write them.
Open Components is part of UXFront, written and maintained by @alexgrozav.
Homepage
·
Documentation
·
Changelog
·
llms.txt
·
Issue Tracker
Table of contents
- The standard
- Reading the standard
- For agents
- Bugs and feature requests
- Contributing
- Creator
- Copyright and license
The standard
The standard has three layers, one for each audience a component serves. A component meets the standard when it meets all three.
- User experience (UX): components behave the way people expect them to, with any input, on any device and for every ability. Accessible, predictable, every state, adaptive.
- Developer experience (DX): one clear API, learned once and used everywhere. Knowing one component means knowing them all. Consistent, type-safe, composable, controllable.
- Agentic experience (AX): components that AI agents can read, reason about and build with, so their output meets the same bar as yours. Semantic, described, deterministic, verifiable.
Learn the guidelines once, then hold every component to the same high standards, whether you or your agents write it.
Reading the standard
Start with the Introduction, which walks through the three
layers and what each one asks of a component. The guidelines live as markdown in
content/docs/, so you can also read them right here on GitHub.
For agents
If your agent supports MCP, connect it to our MCP server at https://mcp.opencomponents.dev/mcp,
free and with no sign-in. It gives agents the contracts, every rule and the docs a section at a
time, as tools. The MCP Server page
shows how to connect Claude, Codex, Cursor, VS Code, Gemini CLI, Grok, Hermes Agent and other
clients.
Or install our agent plugin, in plugins/open-components, which
connects the same server along with skills that build and review components the way the standard
asks. It's an Agent Plugin, so it works in VS Code, GitHub Copilot,
Cursor, Codex, Claude Code and other clients that support the format. The
Agent Plugin page shows how to
install it.
Every page of the documentation is also published as markdown, so agents can read the standard and check their work against it:
https://opencomponents.dev/raw/<path>.mdholds one page, as in/raw/docs/getting-started/introduction.mdfor the introduction.https://opencomponents.dev/raw/<path>.yamlholds a page's contract, with every rule from its checklist, as in/raw/docs/components/button.yaml. It's every requirement in a fraction of the page's length./schemas/contract.jsonis the JSON Schema every contract follows, so you can validate them, and contracts of your own. Each contract names it on its first line, so editors that use the YAML language server check it as you type./llms.txtlists every page and every contract./llms-full.txtholds them all in one file. Point your agent at it to give it the whole standard in one request.
Bugs and feature requests
Found a bug on the site, a gap in the guidelines or an idea for a new one? Please first search for existing and closed issues. If your problem or idea is not addressed yet, please open a new issue.
Contributing
Please read through our contributing guide. There you can find how to run the site locally, edit the guidelines and deploy. Everyone taking part in the project agrees to follow our Code of Conduct.
Thanks goes to these wonderful people!
Creator
Alex Grozav
If you use Open Components in your daily work and feel that it has made your life easier, please consider sponsoring me on GitHub Sponsors. 💖
Copyright and license
Copyright © 2026 UXFront. The site code is released under the
MIT License. The guidelines in content/ and the agent plugin in
plugins/open-components/ are released under the
Creative Commons Attribution 4.0 License: you can share and adapt them,
commercially too, as long as you credit Open Components.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.