Milkey Documentation

Cursor

Using Milkey with Cursor

Milkey brings up-to-date library skills and architecture standards directly into Cursor. Instead of getting outdated code examples from training data, you get your currently curated standards driven seamlessly via MCP.

Installation

You can configure Milkey globally across Cursor or per-project. First, open your Cursor MCP settings.

  1. Go to Settings...Cursor SettingsMCP in the Cursor app.
  2. Click Add a Custom MCP Server, or alternatively edit your ~/.cursor/mcp.json file directly.
~/.cursor/mcp.json
{
  "mcpServers": {
    "milkey": {
      "url": "https://mcp.vexelityai.com/mcp",
      "headers": {
        "MILKEY_API_KEY": "mk_live_your_secret_key_here"
      }
    }
  }
}
Project-Level Config
If you want the rules to apply strictly to one project, you can place this config inside .cursor/mcp.json in your project workspace directory instead of the global application settings.

Setting Up Rules

To make standard usage automatic without needing to repeatedly prompt it, it's highly recommended to add a Cursor Rule in .cursorrules telling it to query the Milkey MCP.

.cursorrules
Always use the Milkey MCP server when I ask about skills, coding standards, or architecture workflows.

Using Milkey

In the Cursor Chat or Composer, you can manually trigger pulling down a specific setup you have connected:

Prompt Example
Use milkey to show me how we handle data transformations in this project.
Use milkey to write the Tailwind components based on our standard UI guidelines.