MCPs ClaudeCursor Foundations

What is MCP?

Model Context Protocol basics and when to use MCP servers

31 of 66

Model Context Protocol

MCP (Model Context Protocol) is a standard for connecting AI agents to external tools and data sources. Think of it as a plugin system for AI.

Core concepts

  • MCP Server: a process that exposes tools and resources to the agent
  • Tools: functions the agent can call (search, navigate, analyze)
  • Resources: read-only data the agent can access (docs, schemas)

Common MCP servers

ServerPurpose
Context7Fetch up-to-date library documentation
PlaywrightBrowser automation and testing
File systemEnhanced file operations
DatabaseQuery databases directly

When to use MCPs vs Skills

Use MCPs when...Use Skills when...
You need live external dataYou need reusable instructions
The tool requires a running processThe task is pure text/code generation
Integration with APIs or browsersFollowing a specific workflow pattern

Key takeaway

MCPs extend what an agent *can do* (new capabilities). Skills guide *how* the agent works (new behaviors). Start with skills; add MCPs when you need external integrations.

Source: modelcontextprotocol.io. The full MCP specification and server registry.

Working out which model to run this on? See The Codex. Packaging it as a reusable skill? See The Armory.