glossary
What is MCP (Model Context Protocol)?
MCP is the open protocol that lets AI agents like Claude Code read and act on tools such as a to-do list. Definition and what a Doozy MCP server exposes.
MCP (Model Context Protocol) is an open standard that lets AI tools connect to outside systems. Instead of a separate plugin for every app and every AI assistant, one MCP server works in any compatible client: Claude Code, Cursor, Codex, Claude Desktop, and more.
What it means
Think of MCP like a USB port for AI. A server exposes a set of tools (like "list to-dos" or "create a task"). Any AI client that speaks MCP can discover those tools and use them. You write the server once, and it works everywhere.
This matters because without MCP, every AI tool needs its own integration for every app. With MCP, one server covers all the clients. That means Doozy's MCP server gives the same tools to Claude Code, Cursor, Codex, and any other MCP client, with one URL.
Example
You connect Claude Code to Doozy's MCP server. Now you can ask Claude Code to "list my active to-dos" and it reads them from Doozy. Create a task from a code review, close it when the fix ships, hand research to a Doozy agent. The same server works if you switch to Cursor or any other MCP client.
Where it appears in Doozy
Doozy runs a hosted MCP server at `https://api.usedoozy.com/mcp`. Nothing to install, nothing to run. It gives any connected client access to to-dos, chats, captures, agents, and skills in your workspace, with permissions you control per key.