The Analyse MCP server
Analyse ships a native, remote MCP server. Connect Claude, Codex, Cursor, or any client that speaks the Model Context Protocol and query your analytics in plain language.
Open standard, zero hosting
A remote MCP server over HTTP with bearer-token auth — nothing to install or host. If your client speaks MCP, it can connect.
Analytics as tools
Query events, build funnels, list cohorts, create segments, read your analytics overview, and search your knowledge base — exposed as typed MCP tools.
Scoped API keys
Every key is tied to the sites you choose when you create it, and can be revoked at any time from your settings.
Connected in a minute
Create a key, paste one config block into your client, done.
- 1
Create an API key
In Analyse, open your account's MCP settings, create an API key, and choose which sites it can access.
- 2
Add the server to your client
Paste the Analyse MCP URL into your client's MCP configuration with your key as a bearer token — the settings page shows a ready-made snippet.
- 3
Ask anything
Your assistant now has live analytics tools — ask about drop-off, retention, or top sources and get grounded answers.
Frequently asked questions
What is MCP?
The Model Context Protocol is an open standard that lets AI assistants securely connect to external tools and data. Analyse implements it as a remote server, so assistants can query your analytics directly.
Which clients can connect?
Any MCP client: Claude (claude.ai, Desktop, and Claude Code), OpenAI Codex, Cursor, Windsurf, Zed, VS Code with MCP support, or your own agent built on an MCP SDK.
Is the MCP server included in my plan?
The MCP server is available on the Pro plan and above. Every plan starts with a free trial, so you can try it before committing.