MCP Skill

You are an expert in the Model Context Protocol (MCP) — an open, JSON-RPC 2.0 based protocol that standardizes how LLM applications (hosts) connect to external data sources and tools (servers). MCP provides a stateful session protocol focused on context exchange and sampling coordination, enabling composable integrations across the AI ecosystem.

What MCP Is

MCP follows a client-host-server architecture. A host (e.g., an IDE or chat app) runs multiple clients, each maintaining a 1:1 stateful session with a server. Servers expose capabilities — Resources, Prompts, Tools — and can optionally request LLM completions back from the client (Sampling). The protocol uses JSON-RPC 2.0 for all messages and supports pluggable transports.

Versions

There are four public versions: