MCP

The Model Context Protocol (MCP) is an open standard that allows AI assistants like Claude, Cursor, and other AI-powered tools to securely connect with external data sources and systems. Think of it as a "universal remote" that lets your AI tools access real-world data and functionality.

Ultracite supports MCP to supercharge your AI development workflow.

Installation Guide

Step 1: Choose Your AI Tool

First, make sure you're using an AI development tool that supports MCP:

• Claude Desktop (Free - recommended for beginners)
• Cursor (AI-powered code editor)
• Windsurf by Codeium (AI development platform)
• Other MCP-compatible tools

Step 2: Locate Your Configuration File

Depending on your AI tool, you'll need to edit one of these files:

• Claude Desktop: ~/Library/Application\ Support/Claude/claude_desktop_config.json for macOS and %APPDATA%\Claude\claude_desktop_config.json for Windows
• Cursor: .cursor/mcp.json
• Windsurf: .codeium/windsurf/mcp_config.json
• Other tools: Check your tool's MCP documentation

Step 3: Add Ultracite Configuration

Copy and paste this configuration into your MCP config file:

{
  "mcpServers": {
    "ultracite": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://www.ultracite.ai/api/mcp/mcp"
      ]
    }
  }
}

Step 4: Restart Your AI Tool

Close and reopen your AI application for the changes to take effect.

Step 5: Verify the Connection

Test the integration by asking your AI assistant:

What Ultracite rules are available?

If successful, your AI should be able to list and explain Ultracite rules!

Multiple MCP Servers

You can use Ultracite alongside other MCP servers:

{
  "mcpServers": {
    "ultracite": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://www.ultracite.ai/api/mcp/mcp"]
    },
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"]
    },
    "filesystem": {
      "command": "npx", 
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/project"]
    }
  }
}

Usage Examples

Exploring Available Rules

Ask your AI assistant:

What rules does Ultracite have for React hooks?

Expected response will include relevant React hook linting rules and their descriptions.

Understanding Rule Categories

Show me all accessibility rules in Ultracite

Your AI can list and explain the a11y rules configured in Ultracite.

Comparing Presets

What's the difference between ultracite/core and ultracite/react?

Get detailed information about different preset configurations and when to use them.

Security and Privacy

Data Handling

• The Ultracite MCP server only provides public linting rule documentation and configuration information
• No personal data or code is transmitted to our servers
• All communication happens through your chosen AI tool's security layer

Authentication

• No authentication required for public documentation access
• Future premium features may require API keys
• Always use official Ultracite MCP endpoints