AI Agent ArchitectureMCP, AI Skills & Agent Design Patterns

The complete guide to how AI agents discover and orchestrate Skills, Tools, and Protocols — with practical patterns for Claude Code, Cursor, and Cline.

Get Started

What is MCP?

What are AI Skills?

View on GitHub

🧠

Concepts & Vision

Why "reliable reference sources" matter — the core philosophy behind AI-driven development and how to overcome AI's fundamental limitations.

Read the Vision

🔌

MCP (Model Context Protocol)

The standard protocol by Anthropic for connecting AI agents to external tools, APIs, and data sources. Learn MCP servers, clients, and hosts.

Learn about MCP

📋

AI Skills (Domain Knowledge)

Reusable Markdown-based knowledge files (skill.md) that give AI agents specialized capabilities. Works with npx skills, Claude Code, Cursor, and Cline.

Explore AI Skills

🤖

Agents & A2A

Sub-agents, orchestration patterns, and the Agent-to-Agent protocol — how autonomous agents collaborate.

Discover Agents

🏗️

Three-Layer Architecture

The Agent / Skills / MCP three-layer model and how they compose into production-ready AI agent systems.

See Architecture

🗺️

Strategy & Roadmap

Build priorities, composition patterns, and practical roadmaps for MCP and Skill construction.

View Strategy

Note: This documentation reflects the author's practical insights gained through building and operating AI agent systems with Claude. It is not official documentation from Anthropic or any other organization. Contributions and discussions via GitHub Issues are welcome.