@xlambda-tech/postgres

Provision and manage xlambda-managed PostgreSQL databases.

import { PostgresClient } from '@xlambda-tech/postgres';

const client = new PostgresClient({ apiKey: process.env.XLAMBDA_API_KEY! });

const db = await client.databases.create({ projectId, name: 'my-app-db' });
console.log(db.connectionString); // shown once - store it now

const pool = await db.getPool(); // configured pg.Pool, ready to use
const { rows } = await pool.query('select 1');

getPool() uses the PgBouncer/transaction-pooled port (6432) — the right default for app steady-state traffic. Use getSessionPool() (port 6433, direct session) for migrations, DDL, advisory locks, or anything needing prepared statements, which transaction pooling breaks.

pg is an optional peer dependency — only required if you call getPool()/getSessionPool(). Everything else (provisioning, users, backups) has no dependency on it.

Sub-users and backups

const readonly = await client.databases.users(db.id).create({ permission: 'readonly' });
const backup = await client.databases.backups(db.id).create();

Not covered — use the dashboard

The SQL console (arbitrary query execution, schema introspection, saved queries, table browser CRUD) is deliberately not wrapped here — it's built for the web dashboard's UI, not for calling from your own backend code. A real application should connect with getPool()/getSessionPool() and use pg (or an ORM) directly, the same as it would against any other Postgres database.