OriginUI MCP Server
Enables searching, browsing, and installing OriginUI components through the Model Context Protocol. It provides detailed component information, visual previews, and installation commands compatible with the shadcn CLI.
README
OriginUI MCP Server
A Model Context Protocol (MCP) server that provides access to OriginUI components, enabling Claude Code agents to search, browse, and install OriginUI components seamlessly.
Features
- Component Search: Search components by name, category, or tags
- Component Details: Get comprehensive information about specific components
- Installation Commands: Generate proper
pnpm dlx shadcn@latest addcommands - Visual Previews: Get component screenshots and visual descriptions to assess fit
- Component Preview: View styling information and usage examples
- Category Filtering: Browse components by category (buttons, inputs, forms, etc.)
🚀 Quick Start
TL;DR: Add this to your ~/.claude-code/mcp.json file:
{
"mcpServers": {
"origin-ui": {
"command": "npx",
"args": ["--yes", "github:kelvinchng/origin-ui-mcp"],
"env": {}
}
}
}
Restart Claude Code and ask: "What OriginUI button components are available?"
Installation
Option 1: Use with npx (Recommended - No Cloning Required!)
No installation needed! You can use the MCP server directly from the GitHub repository with npx.
Option 2: Clone and Build Locally
git clone https://github.com/kelvinchng/origin-ui-mcp.git
cd origin-ui-mcp
npm install
npm run build
Option 3: Global Installation (Coming Soon)
npm install -g origin-ui-mcp
Claude Code Integration
Step 1: Add to Claude Code MCP Configuration
Create or edit your Claude Code MCP configuration file at ~/.claude-code/mcp.json:
Option A: Using npx (Recommended)
{
"mcpServers": {
"origin-ui": {
"command": "npx",
"args": ["--yes", "github:kelvinchng/origin-ui-mcp"],
"env": {}
}
}
}
Option B: Using local clone
{
"mcpServers": {
"origin-ui": {
"command": "node",
"args": ["/absolute/path/to/origin-ui-mcp/dist/index.js"],
"env": {}
}
}
}
Note: With Option A (npx), no manual installation or path configuration is needed!
Step 2: Restart Claude Code
After adding the MCP configuration, restart Claude Code to load the OriginUI MCP server.
Step 3: Verify Integration
You can verify the integration is working by asking Claude Code:
"What OriginUI components are available for buttons?"
Claude Code should now be able to search, describe, and provide installation commands for OriginUI components.
Usage Examples
Once integrated with Claude Code, you can ask questions like:
- "Find me payment-related components"
- "How do I install the button component?"
- "What styling options does the card component have?"
- "Show me all navigation components"
- "I need a component for user input forms"
Claude Code will use the MCP server to provide accurate information and installation commands.
📸 Visual Component Assessment
One of the key features is the ability to get visual previews of components to help LLMs make better recommendations:
"Show me what the payment component looks like"
"Get a screenshot of the button component"
"I want to see the visual style of navigation components"
The MCP server will provide:
- Visual descriptions of component appearance
- Links to live examples on OriginUI
- Theme-specific information (light/dark modes)
- Mobile responsiveness details
- Use case recommendations based on visual style
Available Tools
search_components
Search for OriginUI components by name, category, or tags.
Parameters:
query(string, required): Search querycategory(string, optional): Filter by categorylimit(number, optional): Maximum results (default: 10)
Example:
search_components({
query: "payment",
category: "form",
limit: 5
})
get_component_details
Get detailed information about a specific component.
Parameters:
componentId(string, required): Component ID (e.g., "comp-163")
Example:
get_component_details({
componentId: "comp-163"
})
list_components
List all available components with basic information.
Parameters:
category(string, optional): Filter by categorylimit(number, optional): Maximum results (default: 50)
get_install_command
Get the installation command for a specific component.
Parameters:
componentId(string, required): Component ID
Returns:
pnpm dlx shadcn@latest add https://originui.com/r/comp-163.json
get_component_preview
Get component preview with styling information and usage examples.
Parameters:
componentId(string, required): Component ID
get_component_screenshot
Get component visual preview to help assess if it fits your project.
Parameters:
componentId(string, required): Component IDtheme(string, optional): Theme preference ("light", "dark", "both")
Returns:
- Visual descriptions of component appearance
- Screenshots when available
- Links to live examples
- Use case recommendations
- Installation instructions
Component Categories
button- Button componentsinput- Input and form controlsselect- Select and dropdown componentsnavbar- Navigation componentscard- Card and container componentsform- Form-related componentslayout- Layout componentsnavigation- Navigation elementsfeedback- Feedback and status componentsdata-display- Data display componentsoverlay- Modal and overlay componentstypography- Text and typography componentsmedia- Media components
Example Usage with Claude Code
// Search for button components
const buttons = await search_components({
query: "button",
category: "button",
limit: 10
});
// Get details for a specific component
const details = await get_component_details({
componentId: "comp-163"
});
// Get installation command
const installCmd = await get_install_command({
componentId: "comp-163"
});
// Get visual preview to assess fit
const screenshot = await get_component_screenshot({
componentId: "comp-163",
theme: "both"
});
// Install the component
// pnpm dlx shadcn@latest add https://originui.com/r/comp-163.json
Development
# Install dependencies
npm install
# Build the project
npm run build
# Run in development mode
npm run dev
# Start the MCP server
npm start
How It Works
- Component Discovery: The server maintains a registry of OriginUI components with metadata
- Dynamic Fetching: Component details are fetched from OriginUI's JSON registry on demand
- Caching: Components are cached for 5 minutes to improve performance
- Installation: Generates proper
pnpm dlx shadcn@latest addcommands with correct URLs
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
License
MIT License - see LICENSE file for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.