Raw SQL

Raw SQL drops every abstraction. The @bionicjs/sql plugin hands you the database driver directly — pg, better-sqlite3, or mysql2 — so you get full control over queries, connections, and pooling.

Compose the plugin

Add @bionicjs/sql under the database key in bionicjs.config.ts.

// bionicjs.config.ts
import { defineConfig } from "@bionicjs/core";
import { sql } from "@bionicjs/sql";

export default defineConfig({
  database: sql({ provider: "sqlite", url: "file:./db.sqlite" }),
});

Environment variables

• DATABASE_URL — the single connection string; the .env.example names it.

Using the client

The plugin exposes a raw driver instance as the db export on @bionicjs/core/server:

// server/api/users.ts
import { db } from "@bionicjs/core/server";

export default defineEventHandler(async () => {
  const rows = await db.query("SELECT * FROM users");
  return rows;
});

When to choose raw SQL

• You already write SQL and do not want a query builder or ORM in the way.
• You need driver-specific features or advanced connection pooling.
• Your team values absolute control over generated SQL.

You manage connections and migrations yourself.

Good to know

• The db boundary on @bionicjs/core/server stays the same whether you pick raw SQL, Prisma, Drizzle, or Kysely — only the query files and the config change.
• The client is constructed on the server and never imported into the client bundle.