Model Context Protocol

The Model Context Protocol (MCP) allows AI tools to access prompt-kit components directly. This enables you to browse, preview, and add components without leaving your coding environment.

What is MCP?

MCP is an open protocol that standardizes how applications provide context to LLMs. Think of it as a plugin system that connects your AI tools to external data sources and tools.

Learn more about MCP:

Setup

The setup process depends on your AI tool.

Cursor

  1. Create a .cursor/mcp.json file in your project root
  2. Add the prompt-kit MCP server configuration
{
  "mcpServers": {
      "prompt-kit": {
      "description": "prompt-kit registry",
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "shadcn@canary", "registry:mcp"],
      "env": {
          "REGISTRY_URL": "https://www.prompt-kit.com/c/registry.json"
          }
      }
  }
}
  1. Restart Cursor

Other AI Tools

For other AI tools that support MCP, refer to their specific documentation on how to configure MCP servers. Use the same server configuration above, adapting the file location as needed.

Usage

Once configured, you can start using the prompt-kit components in your AI tool's chat.

Browse Components

Show me all available prompt-kit components

Add a Specific Component

Add the PromptInput component from prompt-kit to my project

Get Component Details

Show me the documentation and preview for the PromptTextarea component

Build a basic chat app

Build a chat app with prompt-kit using: ChatContainer with ScrollButton inside, plus Message and PromptInput
InstallationPrompt Input