Skip to content
alleo wiki Homewiki
/MCP (Model Context Protocol) — The Standard for Connecting AI Tools
📘Concept

MCP (Model Context Protocol) — The Standard for Connecting AI Tools

최종 업데이트:

Definition

MCP (Model Context Protocol) is an open standard protocol led by Anthropic that connects AI models with external tools and data sources.

TL;DR

MCP standardizes how AI models access external services (databases, APIs, file systems, and so on). Just as USB-C connects diverse devices through a single port, MCP lets diverse AI models access diverse tools in a consistent way. From an AEO perspective, MCP expands the paths through which AI can directly access a site's data.

The Structure of MCP

MCP follows a client-server architecture.

  • MCP host: The AI application (Claude Desktop, IDE, etc.)
  • MCP client: The module within the host that communicates with the server
  • MCP server: The service that provides external tools and data to the AI

An MCP server provides three things:

  1. Resources: Data sources (files, DBs, API responses)
  2. Tools: Functions the AI can call (search, computation, API calls)
  3. Prompts: Predefined prompt templates

Implications from an AEO Perspective

As MCP spreads, the AI gains one more path to access a site's data, in addition to the existing crawling and web search paths.

  • Services that provide an MCP server: The AI can directly access structured data
  • Services that provide an API: They can join the AI tool ecosystem via an MCP adapter
  • Content sites: For now, the crawling and web search paths are the main channels, but providing content via MCP will also be possible in the future

The Relationship Between robots.txt, llms.txt, and MCP

File/ProtocolRoleDirection
robots.txtControls bot access to the siteBlock/Allow
llms.txtGuides the AI on the site structureProviding information
MCPProvides tools and data directly to the AIInteraction

The three are complementary. The most comprehensive AI accessibility strategy is to allow crawling with robots.txt, guide structure with llms.txt, and directly provide structured data via MCP.

Frequently Asked Questions

Q. Should ordinary websites also build an MCP server?
A. It is unnecessary for now. MCP is mainly used in development tools, SaaS, and data services. For ordinary content sites, robots.txt + llms.txt + structured data is sufficient.

References

관련 항목

📙How-to
llms.txt Writing Guide
llms.txt is a markdown-format metadata file that helps LLMs efficiently understand site content efficiently, placed at the site root (/) as an AI-friendly site guide.
📘Concept
Agentic RAG — AI Agent-Based Retrieval-Augmented Generation
Agentic RAG is a RAG pattern in which an AI agent evaluates search results and autonomously performs additional searches and tool calls.
📘Concept
AI Agent SEO — Optimization for the Era of Agents Browsing the Web
AI Agent SEO is a strategy for optimizing a site in an environment where AI agents such as ChatGPT Operator and Claude Computer Use directly browse and operate the website.
📘Concept
How RAG Works
RAG is a core technology that combines retrieval and generation to improve AI answer accuracy.
📙How-to
Claude Citation Optimization
Claude citation optimization is the work of optimizing content so Anthropic Claude cites it as a source for its answers.
📙How-to
How to Allow AI Bots in robots.txt
Allowing AI bots means explicitly permitting major AI crawlers such as GPTBot, ClaudeBot, and PerplexityBot to access your site in robots.txt, exposing your content for citation in generative AI answers.

이런 항목도 있어요

📙How-to
How to Allow AI Bots in robots.txt
Allowing AI bots means explicitly permitting major AI crawlers such as GPTBot, ClaudeBot, and PerplexityBot to access your site in robots.txt, exposing your content for citation in generative AI answers.
📘Concept
Agentic RAG — AI Agent-Based Retrieval-Augmented Generation
Agentic RAG is a RAG pattern in which an AI agent evaluates search results and autonomously performs additional searches and tool calls.
📘Concept
AI Agent SEO — Optimization for the Era of Agents Browsing the Web
AI Agent SEO is a strategy for optimizing a site in an environment where AI agents such as ChatGPT Operator and Claude Computer Use directly browse and operate the website.
📘Concept
How RAG Works
RAG is a core technology that combines retrieval and generation to improve AI answer accuracy.
이 페이지가 도움이 됐나요?