Analysis updated 2026-05-18
Ask an AI assistant questions about your website traffic instead of digging through a dashboard.
Generate funnel, retention, and revenue reports for your site through natural conversation.
Let an AI client manage a self-hosted Umami instance's users with explicit opt-in permissions.
| matious89pl/umami-analytics-mcp | 0xradioac7iv/tempfs | 7vignesh/pgpulse | |
|---|---|---|---|
| Stars | 0 | 0 | 0 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | easy | moderate | moderate |
| Complexity | 3/5 | 3/5 | 4/5 |
| Audience | developer | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
Zero-install via npx for Umami Cloud, self-hosting via Vercel or Docker needs a bit more setup.
This project connects Umami, a website analytics tool that tracks visitor traffic, to AI assistants like Claude through something called the Model Context Protocol, or MCP. MCP is a standard way for AI chat tools to plug into outside services and pull in live data or perform actions, rather than only knowing what was in their training. With this connected, you can ask an AI assistant questions about your website's traffic, generate reports, and even manage your Umami account, all through normal conversation instead of clicking around a dashboard. It works with both Umami Cloud, the hosted version, and self-hosted Umami installations that you run yourself. Getting started needs no installation at all: you point your AI client at a single command, supply an API key or login details as environment variables, which are just settings kept outside the code, and it connects. There is also an option to deploy it yourself on Vercel, a hosting platform, or run it in a Docker container, a way of packaging software to run consistently anywhere. Security is a major focus of this project. By default, the assistant can only read data, not change anything: viewing statistics, generating reports on things like funnels, user retention, and revenue, and browsing sessions. If you want the AI to also create or update things, or manage self-hosted user accounts, you have to explicitly turn those abilities on through separate settings. Anything destructive, like deleting a website or resetting data, needs an extra opt-in on top of that. Login secrets are kept only in environment variables and are actively scrubbed from logs and error messages so they never leak out accidentally. Overall this is a tool for developers and site owners who already use Umami and want to interact with their analytics data through an AI assistant rather than a traditional dashboard. It requires some comfort with environment variables and, for self-hosting, basic server or Docker knowledge. The project is open source under the MIT license.
An MCP server that connects Umami website analytics to AI assistants like Claude, letting you query traffic reports and manage your account through chat, read-only by default.
Mainly TypeScript. The stack also includes TypeScript, Model Context Protocol, Vercel.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.