Jpisnice/shadcn-ui-mcp-server
A mcp server to allow LLMS gain context about shadcn ui component structure,usage and installation,compaitable with react,svelte 5,vue & React Native
What it solves
This project provides a way for AI assistants to directly access and integrate shadcn/ui v4 components into development workflows. It eliminates the need for the AI to guess or hallucinate component code by providing direct access to the latest source code, demos, and metadata for multiple frameworks.
How it works
It is implemented as a Model Context Protocol (MCP) server. It connects to GitHub to retrieve the latest TypeScript source code, blocks, and implementation patterns for shadcn/ui. It supports multiple transport modes (stdio and SSE) and can be deployed via Docker for production use.
Who it’s for
Developers using AI-powered coding assistants (like Claude Desktop, Cursor, or VS Code) who want to build user interfaces using the shadcn/ui ecosystem across different frameworks.
Highlights
- Multi-Framework Support: Provides implementations for React, Svelte, Vue, and React Native.
- Comprehensive Access: Retrieves source code, component demos, and full blocks (e.g., dashboards and forms).
- Flexible Deployment: Supports standard input/output, Server-Sent Events (SSE) for multi-client use, and Docker containerization.
- Smart Caching: Includes GitHub API integration with rate limit handling to ensure efficient data retrieval.
Related
- Project
- Project
- Project
- Project
- Project