Cloudflare logo

Cloudflare

Deploy and manage resources on the Cloudflare developer platform, including Workers, Pages, R2 storage, D1 databases, and more.

How to Use Model Context Protocol (MCP) in Cursor

What is MCP?

Model Context Protocol (MCP) is an open protocol that allows you to provide custom tools to agentic LLMs (Large Language Models) in Cursor's Composer feature.

Installation Steps

  1. Open Cursor Settings
    • Navigate to Cursor Settings > Features > MCP
    • Click the "+ Add New MCP Server" button
  2. Configure the Server
    • Name: Give your server a nickname
    • Type: Select the transport type (stdio or sse)
    • Command/URL: Enter either:
      • For SSE servers: The URL of the SSE endpoint
      • For stdio servers: A valid shell command to run the server

Example Configurations

For stdio Server (Cloudflare Example):

Command: node ~/mcp-quickstart/cloudflare-server/build/index.js --api-token $CLOUDFLARE_API_TOKEN --account-id $CLOUDFLARE_ACCOUNT_ID

For SSE Server:

URL: https://api.cloudflare.com/client-v4/sse

Using MCP Tools

Tool Availability

  • • After adding a server, it will appear in your MCP servers list
  • • You may need to click the refresh button to populate the tool list

Using Tools in Composer

  • • The Composer Agent automatically uses MCP tools when relevant
  • • You can explicitly prompt tool usage by:
    • Referring to the tool by name
    • Describing the tool's function

Tool Execution Process

  • • Displays a message in chat requesting approval
  • • Shows tool call arguments (expandable)
  • • Executes the tool upon user approval
  • • Displays the tool's response in the chat

Important Notes

  • • MCP tools may not work with all models
  • • MCP tools are only available to the Agent in Composer
  • • For Cloudflare integration, you need to set up CLOUDFLARE_API_TOKEN and CLOUDFLARE_ACCOUNT_ID environment variables
  • • Create an API token from the Cloudflare dashboard with appropriate permissions
  • • Get your Account ID from the Cloudflare dashboard overview
  • • For security, use a wrapper script that sets the environment variables before running the server
  • • Use scoped API tokens with minimum required permissions
  • • Be mindful of API rate limits and usage quotas
  • • Implement proper error handling for API responses
  • • Consider using environment-specific tokens for different stages
  • • Monitor Workers usage and resource consumption
  • • Follow deployment best practices for Workers and Pages
  • • Implement proper backup strategies for R2 and D1 data
  • • Use appropriate caching strategies for optimal performance