
## The Problem

Sometimes your database already contains schema changes that are missing from your Prisma migration files.

In these situations:

- the actual database structure is ahead
- but Prisma migration history is behind

---

## The Goal

You want Prisma migrations to catch up with the current database state.

Instead of recreating everything from scratch, you can generate a migration representing the difference between:

- the current database
- and your current `schema.prisma`

---

## Step 1 — Create the Migration Folder

```bash
mkdir prisma/migrations/20260515000000_add_file_type
```

---

## Step 2 — Generate the Migration SQL

```bash
npx prisma migrate diff \
  --from-config-datasource \
  --to-schema prisma/schema.prisma \
  --script > prisma/migrations/20260515000000_add_file_type/migration.sql
```

This generates SQL representing the difference between:

- the configured datasource database
- and your Prisma schema

---

## Important Reminder

Always inspect the generated SQL before applying or marking migrations.

Prisma will generate SQL based purely on detected differences.

So if your schema contains accidental changes, those will also appear in the generated migration.

---

## Final Thoughts

`prisma migrate diff` is one of Prisma's most useful recovery tools.

Especially when working with existing databases or repairing migration drift.

Understanding the difference between:

- database state
- Prisma schema
- migration history

makes Prisma migrations much easier to reason about.


Product Journey