MCP
Model Context Protocol — the USB port for AI tools.
What is MCP?
The Model Context Protocol is an open standard introduced by Anthropic that provides a universal way for AI agents to connect to external tools and data sources. Think of it as USB for AI — a single interface that any tool can implement and any agent can use.
Before MCP, every tool integration required custom code: API wrappers, authentication handling, response parsing. MCP replaces all of that with a single protocol.
How it works
The protocol defines three roles:
- Server — exposes tools (Membot is an MCP server)
- Client — the AI agent that discovers and calls tools
- Transport — the communication channel between server and client
The server declares its tools (name, description, parameters). The client discovers them at startup and can call them during conversations. The protocol handles serialization, error handling, and lifecycle management.
Transport modes
Why MCP matters for Membot
Zero integration code. Any MCP-compatible agent can use Membot immediately. No SDK, no wrapper library, no API client. Just point the agent at the server and it discovers the tools.
Composable. Membot can run alongside other MCP servers. An agent might use Membot for memory, a filesystem server for file access, and a web search server for live data — all through the same protocol.
Compatible agents
- Claude Code — Anthropic's CLI agent
- Claude Desktop — Anthropic's desktop app
- OpenClaw — open-source Claude Code fork
- Any agent implementing the MCP client specification
Further reading
modelcontextprotocol.io — official MCP specification
Get started — connect Membot to your agent