What is MCP?
TL;DR
MCP (Model Context Protocol) is an open standard by Anthropic that lets AI agents connect to external tools and services, enabling them to take actions on websites.
Last updated: 2026-03-09
Definition#
Model Context Protocol (MCP) is an open standard developed by Anthropic that defines how AI agents connect to external tools, data, and services. It provides a structured way for AI models to go beyond reading content and actually interact with websites and applications.
MCP uses a client-server architecture. The AI model acts as the client and your service acts as the server. Your MCP server exposes three types of capabilities: tools (functions the AI can call), resources (data the AI can read), and prompts (templates that guide AI behavior).
With MCP, an AI assistant could check product availability on your e-commerce site, book a table at your restaurant, or pull a report from your SaaS dashboard — all without the user visiting your website directly.
Why It Matters for AI Readiness#
MCP is the action layer of the agentic web. While llms.txt lets AI read about your site and NLWeb lets AI ask questions, MCP lets AI do things on your site.
For transactional businesses (e-commerce, booking, SaaS), MCP will become increasingly important as users rely on AI agents to act on their behalf. See our MCP Protocol page for more details.
Related Concepts#
MCP builds on LLMs and enables the agentic web. It is complementary to llms.txt (reading) and NLWeb (questioning).
Related Pages
Was this page helpful?