PerSQL on Railway — starter

A minimal Node web service backed by an isolated PerSQL SQLite database. Each page load writes a row and reads the count — that round trip is the whole integration.

Deploy

1. Get a database. Visit railway.persql.com/connect, sign in, and provision a database. You'll get three values:

   Var	What
   PERSQL_API_URL	https://api.persql.com
   PERSQL_DATABASE	<namespace>/<db-slug>
   PERSQL_TOKEN	a token scoped to that one database

2. Deploy. Click the button above. Railway prompts for PERSQL_DATABASE and PERSQL_TOKEN — paste the values from step 1 (PERSQL_API_URL is pre-filled). Prefer the dashboard? New Project → Deploy from GitHub repo → persql/railway-starter works too; railway.json pins the build, start command, and healthcheck.

3. Open the service URL. The visit counter increments on each refresh.

How it connects

import { PerSQL } from "@persql/sdk";

const db = new PerSQL({
  token: process.env.PERSQL_TOKEN,
  baseURL: process.env.PERSQL_API_URL, // defaults to https://api.persql.com
}).database(process.env.PERSQL_DATABASE); // "namespace/db-slug"

await db.query("SELECT count(*) AS n FROM visits");

See server.js for the full example and docs.persql.com/integrations/railway for the integration guide, including per-PR preview databases.

Billing

PerSQL is usage-metered against a prepaid balance — no per-database fee. A schema-only database that you don't query costs effectively nothing; you pay for the requests, rows, and storage you actually use.