Why Prisma IDB?

The Problem

IndexedDB is the only real storage API in the browser. It's also one of the worst developer experiences in web development: callback-based, untyped, no query language, no relations, no migration strategy. Every project that touches it ends up building a wrapper.

Most wrappers (Dexie, idb, localForage) give you a friendlier API but stop there. You still have to:

• Define your schema separately from your server schema
• Build your own sync logic when you need online/offline support
• Handle conflicts, authorization, and data ownership yourself
• Maintain type definitions manually

The Solution

Prisma IDB generates a fully-typed IndexedDB client directly from your existing Prisma schema. If you already use Prisma on the server, you define your models once and get both a server client and a browser client with the same API.

// Server (Prisma Client)
const todos = await prisma.todo.findMany({ where: { userId } });

// Browser (Prisma IDB) — same API
const todos = await idb.todo.findMany({ where: { userId } });

What about sync?

This is where most IndexedDB libraries leave you on your own. Prisma IDB includes a complete bidirectional sync engine:

• Outbox pattern for reliable mutation push with retry
• Ownership DAG for structural authorization
• Changelog materialization for efficient, deterministic pulls
• Conflict resolution that doesn't require manual intervention

It's optional — you can use Prisma IDB purely as a local database. But when your app needs to go online, you flip one flag instead of building a sync system.

Good fit for

• Offline-first apps that need to work without network
• Progressive Web Apps with local data and background sync
• Collaborative tools with multi-device data
• Any app already using Prisma that wants browser-side storage

Ready to start? Head to the Quick Start guide.