umami

Umami is presented as a privacy-first analytics platform that brings traffic, campaigns, behavior, conversions and revenue together in one place. The project emphasizes that it does not use cookies or surveillance techniques, and can be operated either self-hosted or through a hosted cloud offering. A public demo is linked from the README, and project support channels include GitHub, Twitter, LinkedIn and Discord. The code is released under the MIT license. Installation from source requires a server with Node.js 18.18 or later and a PostgreSQL database (version 12.14 or later). The documented flow is to clone the repository, install dependencies with pnpm, create an `.env` file containing a `DATABASE_URL` connection string, run the build, then start the application. The build step creates the database tables on a first install and also seeds a login user with the username `admin` and password `umami`. By default the app starts on `http://localhost:3000`, so the documentation notes that you will typically need to proxy requests from a web server or change the port to serve it directly. Several environment options are documented. `API_URL` can change the base URL used for internal UI API calls; relative paths are served under `BASE_PATH`, while absolute URLs are proxied through the local `/api` route. `TWO_FACTOR_ENCRYPTION_KEY`, a 64-character hex string generated for example with `openssl rand -hex 32`, enables two-factor authentication; until it is set, two-factor authentication is unavailable and cannot be required. For containerized deployment, the project publishes Docker images and provides a Docker compose file that runs Umami together with a PostgreSQL database. Updates are handled either by pulling the source, reinstalling dependencies and rebuilding, or by pulling new images and recreating the containers with Docker compose. A detailed getting-started guide is hosted at the project's documentation site.