Prisma ORM Best Practices

prismadatabaseormnode

Best practices for Prisma schema design, queries and migrations

Целевые файлы

Файл	Формат
.cursorrules	plaintext
CLAUDE.md	markdown

Содержимое

Define all models in schema.prisma with explicit @id, @unique, and @index decorators. Use String @id @default(cuid()) for primary keys unless the domain requires integers. Always add createdAt DateTime @default(now()) and updatedAt DateTime @updatedAt fields. Name relation fields clearly: author User, authorId String — always include both. Use prisma migrate dev for development; never edit migration files by hand after creation. Never use prisma db push in production; always use prisma migrate deploy. Instantiate PrismaClient as a singleton; do not create new instances per request. Use select or include explicitly; avoid loading entire models when only a few fields are needed. Wrap multi-step writes in prisma.$transaction() to ensure atomicity. Use Prisma's built-in pagination: take/skip for offset or cursor-based with cursor/take. Seed the database via prisma/seed.ts; run with prisma db seed, not raw SQL scripts. Avoid raw queries (prisma.$queryRaw) unless Prisma's API cannot express the query. Add @@index on all foreign keys and frequently filtered fields. Handle Prisma errors with instanceof PrismaClientKnownRequestError checks.

CLI

npx mindaxis apply prisma-rules --target cursor --scope project

Используется в паках

Не используется ни в одном паке

← Правила