Model Context Protocol (MCP)
How do AI agents use tools?
Why MCP won: Standard interface for creating tools and providing context to LLMs.
Model Context Protocol: MCP defines how AI models access external capabilities — databases, APIs, file systems, and other tools. It's the bridge between the model's reasoning and the world's data.
- MCP Servers — Server implementations
- MCP Clients — Client integration
MCP vs A2A
| Aspect | MCP | A2A |
|---|---|---|
| Focus | Tool use — agent ↔ external capability | Agent ↔ agent communication |
| Scope | How agents access tools | How agents coordinate |
| Analogy | HTTP (accessing resources) | TCP (establishing connections) |
They're complementary: MCP for tool access, A2A for inter-agent coordination.
Introduction
Implementation
Context
- Agent Protocols — The protocol landscape
- A2A Protocol — Agent-to-agent communication
- Agent Commerce — Transaction protocol
- Claude Code
- AI Coding
- Agent Coding