Model Context Protocol (MCP) Documentation
by Anthropic / MCP (Linux Foundation)
The open standard for connecting LLMs and agents to your data, tools, and workflows — build once, integrate everywhere.
Overview
The Model Context Protocol (MCP) is an open-source standard — originally created at Anthropic and now hosted by the Linux Foundation — for connecting AI applications like Claude, ChatGPT, and coding assistants to external systems. The docs describe MCP as a 'USB-C port for AI applications': a single standardized way for a model to reach data sources (files, databases), tools (search, calculators, APIs), and reusable workflows (prompts). The site covers the core client–server architecture and concepts, step-by-step guides to build MCP servers that expose your data and tools, guides to build MCP clients and MCP-powered apps that run inside AI hosts, the versioned protocol specification (2025-11-25), and SDKs across multiple languages. Because MCP is supported across a broad ecosystem — Claude, ChatGPT, VS Code, Cursor, and more — an integration built once works across many hosts, which is why it has become the de-facto plumbing for tool-using agents.
At a Glance
- Topic
- Frameworks
- Level
- Intermediate
- Format
- Documentation
- Cost
- Free
- Duration
- Self-paced reference + build guides
- Provider
- Anthropic / MCP (Linux Foundation)
- Hands-on
- Yes — code/exercises
- Certificate
- None
What You’ll Learn
- ✓How the MCP client–server architecture works and how hosts, clients, and servers communicate
- ✓How to build an MCP server that exposes your own data, tools, and prompts to any MCP-compatible AI app
- ✓How to build MCP clients and apps that connect to and consume MCP servers
- ✓The core MCP primitives — tools, resources, and prompts — and when to use each
- ✓How to read and target the versioned MCP specification for interoperable integrations
Highlights
- •The de-facto open standard for agent/tool integration, backed by Anthropic and hosted by the Linux Foundation
- •Supported across Claude, ChatGPT, VS Code, Cursor and a broad client/server ecosystem — build once, integrate everywhere
- •Official SDKs, build-a-server / build-a-client tutorials, and a public versioned spec
Who It’s For
Best For
- ✓AI engineers building tool-using or agentic applications
- ✓Developers integrating LLMs with internal data sources, APIs, and databases
- ✓Teams standardizing how their AI products connect to external systems
Prerequisites
- •Comfortable with a programming language (e.g. Python or TypeScript)
- •Basic understanding of LLMs, tool/function calling, and client–server APIs
FAQ
What is Model Context Protocol (MCP) Documentation?
The official documentation for the Model Context Protocol (MCP), the open standard that lets AI applications and agents connect to external data sources, tools, and workflows. Aimed at engineers building agentic apps, MCP servers, or MCP clients who need a vendor-neutral integration layer.
Is Model Context Protocol (MCP) Documentation free?
Model Context Protocol (MCP) Documentation is free to access.
What level is Model Context Protocol (MCP) Documentation for?
Model Context Protocol (MCP) Documentation is aimed at a intermediate audience. Recommended background: Comfortable with a programming language (e.g. Python or TypeScript), Basic understanding of LLMs, tool/function calling, and client–server APIs.
How long does Model Context Protocol (MCP) Documentation take?
Expect roughly Self-paced reference + build guides. Most learners work through it at their own pace.
What will I learn from Model Context Protocol (MCP) Documentation?
You'll learn: How the MCP client–server architecture works and how hosts, clients, and servers communicate; How to build an MCP server that exposes your own data, tools, and prompts to any MCP-compatible AI app; How to build MCP clients and apps that connect to and consume MCP servers; The core MCP primitives — tools, resources, and prompts — and when to use each; How to read and target the versioned MCP specification for interoperable integrations.