MCP
Anthropic's open standard for connecting LLMs to external tools and data sources over JSON-RPC.
Definition
The Model Context Protocol (MCP) is an open standard from Anthropic that lets LLM clients (Claude Code, Claude Desktop, Cursor, and others) connect to tool servers and data sources over JSON-RPC. An MCP server exposes a set of tools and resources; an MCP client discovers and calls them. The protocol is transport-agnostic — stdio, HTTP, or WebSocket — and decouples the LLM from per-integration glue code.
When to use
Use MCP when you want a Claude, GPT, or Gemini client to call your tools or read your data without writing a separate integration per LLM. A single MCP server works across every MCP-capable client.
See also
- agent — An LLM-driven program that plans, calls tools, observes results, and loops until it reaches a goal.
- Claude Code — Anthropic's official terminal-based coding agent, with MCP support and a plugin ecosystem.
- n8n — Open-source workflow automation tool with 400+ integrations, self-hostable, fair-code licensed.
Mentioned in
- agency-os: the Notion board that plans your work, then ships it
- Cursor rules not loading: 4 causes and how to fix each
- Fix Cursor 3 "Agent Execution Timed Out" and Jupyter not opening
- How to fix Claude Code 'No available IDEs detected' in WSL with Cursor
- How to fix Claude Code creating a 'nul' file on Windows
- How to fix MCP Token expired without refresh token in Claude Code
- Launching content-distribution-mcp: one finished post, eight channels
- Make vs Zapier in 2026: pricing per task, AI features, and which automation platform to pick
- n8n MCP server: build, lint, and debug workflows from your AI agent
- Notion Custom Agents vs Claude Code: when to use each in 2026
- The n8n Nodes Catalog: a complete, machine-readable reference
- Three Claude skills for end-to-end Ghost blog publishing