---
title: Running the Platform
description: Install prerequisites, configure env, seed the demo, boot server plus register plus docs, and verify with health and login.
---

You go from a clean machine to a working pharmacy in about ten minutes. You need Bun, pinned to 1.4.3 by the `packageManager` field in `package.json`, and Git. The server runs on SQLite through `DATABASE_PATH` by default, and it switches to PostgreSQL only when you set `DATABASE_URL`.

### Check your prerequisites

You confirm the toolchain before cloning so version errors surface early. The pinned bun version lives in the `packageManager` field of the root `package.json`; a mismatched runtime is the most common cause of lockfile churn.

```bash
bun --version
```

If it does not print 1.4.3, install that version before you continue.

### Clone and install

You clone the repo, check out the branch you are working on, and install every workspace at once.

```bash
git clone https://github.com/vernonthedev/pharmaflow.git
cd pharmaflow
bun install
```

### Configure the server environment

You copy `apps/server/.env.example` to `apps/server/.env` and fill in your values. Bun loads `.env` from the working directory when `apps/server/src/index.ts` boots, and the full key list in the example file is read through `readEnv` in `apps/server/src/core/config/env.ts`. `JWT_SECRET` must be at least 32 characters or every boot fails fast.

```bash
cp apps/server/.env.example apps/server/.env
```

The server defaults to SQLite: leave `DATABASE_PATH` unset and `apps/server/src/core/database/db.ts` falls back to `resolveDatabaseFile` in `packages/database/src/paths.ts`, which points at a database file under `packages/database/data/`. Set an absolute `DATABASE_PATH` to use another SQLite file, or point `DATABASE_URL` at a Postgres instance instead. `PORT` serves the API, and `ALLOWED_ORIGINS` already lists the register, the marketing app, and the Tauri console in the example file.

```env
JWT_SECRET=<at-least-32-characters-long-random-string>
PORT=8000
ALLOWED_ORIGINS=http://localhost:3000,http://localhost:3001,tauri://localhost,https://tauri.localhost,http://tauri.localhost
```

### Seed the demo data

You create the demo branches, users, medicines, batches, and the `DEMO10` coupon with the seed script in `packages/database`.

```bash
bun --cwd packages/database seed
```

You now sign in as `admin@demo.pharmaflow`, `pharm1@demo.pharmaflow`, or `pharm2@demo.pharmaflow`, all with password `Demo@1234`. You reset any time by running the same command again.

### Boot server, register, and docs

You run each app in its own terminal. The API serves port 8000, the register serves port 3000, and the docs print their own URL.

```bash
bun run dev:server
```

```bash
bun run --cwd apps/register dev
```

```bash
bun run --cwd apps/knowledgebase dev
```

### Verify health and login

You hit `GET /health` in `apps/server/src/index.ts` for the smoke check, then you log in and store a token for the rest of your session. Router routes mount under the `/api/v1` prefix that `apps/server/src/app.ts` applies with `app.route("/api/v1", apiV1)`, so auth sits at `/api/v1/auth`.

```bash
curl http://localhost:8000/health
```

```bash
TOKEN=$(curl -s -X POST http://localhost:8000/api/v1/auth/login \
  -H "Content-Type: application/json" \
  -d '{"email":"admin@demo.pharmaflow","password":"Demo@1234"}' \
  | python3 -c "import sys,json; print(json.load(sys.stdin)['access_token'])")
curl http://localhost:8000/api/v1/auth/me -H "Authorization: Bearer $TOKEN"
```

### Start the desktop console

You run the Tauri v2 console wrapping the register UI from the repository root:

```bash
bunx -p @tauri-apps/cli tauri dev
```

The shell config in `apps/console/tauri.conf.json` points `devUrl` at `http://localhost:3000` and its `beforeDevCommand` starts the register dev server for you, so you can skip the separate register terminal when you use the console. The Rust backend lives in `apps/console/src`.

### Next steps

Walk your first counter sale in /guides/selling-end-to-end, connect an offline device in /guides/offline-first-flow, and learn the auth model in /platform/authentication-and-users.