Schema rules

3 rules that validate database schema design: primary keys, timestamps, and relation configuration. They run against entity-relation data extracted from Prisma schema files, TypeORM and MikroORM entity decorators, or Drizzle table declarations.

Scope: Schema

Detects entities that have no primary key column.

Why: Entities without primary keys cannot be uniquely identified, breaking lookups, joins, and ORM operations. Every table should have an explicit primary key.

model Product {
  name  String
  price Float
}

Scope: Schema

Detects entities missing createdAt/updatedAt timestamp columns.

Entities whose every column is a primary key or a relation's own column are skipped.

Why: Timestamp columns are essential for auditing, debugging, and data integrity. Without them, there is no way to know when a record was created or last modified.

model Post {
  id    Int    @id @default(autoincrement())
  title String
  body  String
}

Scope: Schema

Detects owning-side relations (ManyToOne/OneToOne) without an explicit onDelete behavior.

Why: Without an explicit onDelete, the database default applies, usually RESTRICT or NO ACTION. Deleting a parent record then fails with a constraint error. An explicit cascade behavior also makes the data model self-documenting.

model Comment {
  id     Int  @id @default(autoincrement())
  post   Post @relation(fields: [postId], references: [id])
  postId Int
}