--- title: CLI description: 'The Analytik command line: create, upgrade, init, migrate, seed, import, menubar and doctor.' --- The package installs a binary named `analytik`. Run it with `npx analytik`, `bunx analytik`, or from your scripts. ```bash npx analytik [options] ``` `-h` or `--help` prints the usage, `-v` or `--version` prints the version. ## Database targets Commands that open a database take `--db `. The value is one of: | Value | Database | | -------------------------------------------- | -------------------------------------------------------------------------- | | `postgres://user:pass@host/db` | Postgres. Needs the `postgres` package installed. | | `sqlite:`, `file:`, a `.db` path | A SQLite file. A path ending in `.db`, `.sqlite` or `.sqlite3` also works. | | `:memory:` | A throwaway in-memory SQLite database. | Cloudflare D1 cannot be reached from the command line. See [migrate](#migrate) for how to load the schema with Wrangler. ## Commands ### create Makes a project that serves the dashboard, the tracker and the API from one deployment. On Cloudflare (the default) it also puts it online: it installs the packages, signs in to Cloudflare if needed, creates the D1 database, deploys the Worker and sets the secrets, then prints the address and a generated dashboard password (kept in `.dev.vars`, which is git-ignored). ```bash npx analytik@latest create my-stats npx analytik@latest create my-stats --name shop-stats --no-deploy npx analytik@latest create my-stats --target vercel ``` `--name` sets the Worker and database name (default: the folder's), `--password` the dashboard password (default: a random one) and `--no-deploy` writes the files only. `--target vercel` writes a Vercel function with Postgres; secrets go to `.env`, and you deploy with `npx vercel`. See [Cloudflare Workers and D1](/deploy/cloudflare-workers-d1). ### upgrade Moves a project to the newest Analytik. Run it in the project folder, or pass the folder. ```bash npx analytik upgrade npx analytik upgrade ./my-stats --no-deploy ``` It installs `analytik@latest` with the package manager the project uses (by lockfile: bun, pnpm, yarn, else npm) and, for a Cloudflare project, runs `wrangler deploy`. For Vercel it tells you to deploy. The database schema updates on the first request afterwards. See [Updating](/deploy/backups-retention-and-upgrades#updating). ### init Prints a starter server entry file to stdout. It does not create any files, so redirect the output if you want to keep it. Pick the platform with `--target worker|vercel|bun|node` (default `worker`). ```bash npx analytik init --target bun > server.ts ``` The reminder about `ADMIN_PASSWORD` and `SESSION_SECRET` goes to stderr, so it does not end up in the file. The server refuses to start when `ADMIN_PASSWORD` is empty or `SESSION_SECRET` is shorter than 32 characters. ### migrate Creates the schema or applies pending migrations, the same as `storage.migrate()`. Safe to run repeatedly. ```bash npx analytik migrate --db "$DATABASE_URL" npx analytik migrate --db ./data/analytik.db ``` For D1, print the schema as SQL and load it with Wrangler: ```bash npx analytik migrate --print-sql sqlite > analytik-schema.sql npx wrangler d1 execute analytik --remote --file analytik-schema.sql ``` `--print-sql sqlite|postgres` writes every migration plus the rows that mark them as applied, so `createAnalytik()` does not run them a second time. `npx analytik migrate --d1 ` does not touch anything: it prints these same two commands for that database name and exits with an error code. By default `createAnalytik()` also applies pending migrations on the first request (`autoMigrate`), so on D1 you can skip this step. ### seed Fills a database with generated demo data, for trying the dashboard. Use it on a local or throwaway database. ```bash npx analytik seed --db ./demo.db --days 30 --seed 1 ``` `--db` is required. `--days` (default 30) is how many days of data to generate and `--seed` (default 1) makes the data repeatable. The command prints the site id and API key it created. ### import Adds the history of another analytics tool to a site: Umami, Simple Analytics, Matomo, GoatCounter, Google Analytics 4 or PostHog. ```bash npx analytik import --from umami --file events.csv --file sessions.csv --file event-data.csv --site SITE_ID --db "$DATABASE_URL" ``` `--from` names the source, `--file` is repeated once per export file, `--site` is the id of an existing site and `--db` the database. The command applies pending migrations, reads every file, and only then writes, so a wrong file stores nothing. Running it twice adds nothing the second time. D1 cannot be reached from the command line; use the import in Settings. What each export must contain is in [Importing data](/guides/importing-data). ### menubar Downloads the newest release of the [Mac menu bar app](/guides/mac-menu-bar), puts it into Applications and opens it. It needs macOS 14 or later. ```bash npx analytik menubar npx analytik menubar --dir ~/Applications --no-open ``` `--dir` installs into another folder and `--no-open` does not start the app afterwards. ### doctor Runs a self-check and prints the result of each check. It exits with a non-zero code when one fails. Run it after a deploy or when events do not appear. ```bash npx analytik doctor --db ./data/analytik.db --url https://analytics.example.com ``` | Option | Check | | -------------------- | ------------------------------------------------------------------------------------------------------------- | | (always) | The runtime: Bun, or Node 22.19 or newer. | | `--db ` | The database answers a query and the schema is at the latest migration. | | `--url ` | `GET /api/health` answers with `ok: true`. A URL that already ends in `/health` is used as it is. |