The Model Context Protocol (MCP) allows AI assistants like Claude, v0, and others to interact with external tools and services. The shadcn MCP server enables AI assistants to browse, search, and install workflow steps from this registry using natural language.
Configure the shadcn MCP server to work with the Workflow Registry:
npx -y @shadcn/mcpThis installs the shadcn MCP server globally so AI assistants can use it.
Add the Workflow Elements URL to your MCP configuration:
{
"mcpServers": {
"shadcn": {
"command": "npx",
"args": ["-y", "@shadcn/mcp"],
"env": {
"SHADCN_REGISTRY_URL": "https://workflow-registry.vercel.app"
}
}
}
}For Claude Desktop, this config file is typically located at:
~/Library/Application Support/Claude/claude_desktop_config.json%APPDATA%\Claude\claude_desktop_config.jsonRestart Claude Desktop or your AI assistant to load the new MCP server configuration.
Once configured, you can ask your AI assistant to interact with Workflow Elements using natural language:
"Show me all workflow steps in the registry"
The AI will list all available workflow steps with their descriptions and categories.
"Find workflow steps for sending Slack messages"
The AI will search the registry and show relevant steps matching your query.
"Install the send-slack-message and send-email workflow steps"
The AI will add the steps to your project and install dependencies automatically.
"Build a workflow that sends a Slack notification when a new user signs up, then sends them a welcome email"
The AI will install the necessary steps and help you compose them into a complete workflow.
Here are some example prompts you can use with AI assistants:
"What notification workflow steps are available?""Install all the Vercel-related workflow steps""Show me workflow steps for working with Google Sheets""Add the GitHub create-issue step to my project""Build a workflow that processes CSV data and uploads results to cloud storage"Describe what you want in plain English instead of remembering exact step names and commands.
AI assistants can quickly search and filter steps based on your requirements.
The AI handles installation, dependency management, and basic configuration automatically.
AI assistants can help you combine multiple steps into complete workflows with best practices.
The shadcn MCP server works with any AI assistant that supports the Model Context Protocol:
Make sure you've restarted your AI assistant after updating the configuration file.
Verify that SHADCN_REGISTRY_URL is set to https://workflow-registry.vercel.app in your MCP configuration.
Ensure you're in a valid Next.js or TypeScript project with a package.json file. The MCP server needs write access to create files.