MCP stands for Model Context Protocol. It's an open standard that lets AI assistants like Claude, ChatGPT, and Cursor call external tools, the same way a browser calls an API.
Without MCP, your AI assistant only knows what you paste into it. With MCP, it can reach out and get live data on its own. Your analytics, your database, your CRM, whatever you connect.
An MCP server exposes a set of tools. Each tool does one thing: fetch a page of analytics, look up a funnel, check traffic sources. Your AI assistant sees the list of available tools and calls the right ones based on what you ask.
You say "why did signups drop this week?" and the assistant calls get_analytics, get_funnel, and get_traffic_sources behind the scenes.
You get the answer, not a tutorial on how to find it.
MCP servers come in two flavours. Local servers run on your machine and connect through your terminal (Claude Code, Cursor). Hosted servers run in the cloud and work everywhere: Claude Desktop, claude.ai, your phone, any client that supports the protocol.
Lodd runs both. A local server for Claude Code (npx -y @lodd/mcp-server)
and a hosted server at api.lodd.dev/mcp that
you can add to any MCP client with one click.
Traditional analytics lives in a dashboard you have to visit. MCP puts that data where you already work. Your coding agent can check traffic while it builds features. Your conversation agent can tell you what's working without you opening a tab.
The data comes to you instead of the other way around.
Ready to try it? Set up Lodd in 2 minutes, or see what it looks like.