prisma-cli-db-pull

by prisma

prisma-cli-db-pull — an installable skill for AI agents, published by prisma/cursor-plugin. npx skills add https://github.com/prisma/cursor-plugin --skill prisma-cli-db-pull Download ZIPGitHub8

prisma db pull

Introspects an existing database and updates your Prisma schema to reflect its structure.

Command

prisma db pull [options]

What It Does

• Connects to your database

• Reads the database schema (tables, columns, relations, indexes)

• Updates schema.prisma with corresponding Prisma models

• For MongoDB, samples data to infer schema

Options

Option Description --force Ignore current Prisma schema file --print Print the introspected Prisma schema to stdout --schema Custom path to your Prisma schema --config Custom path to your Prisma config file --url Override the datasource URL from the Prisma config file --composite-type-depth Specify the depth for introspecting composite types (default: -1 for infinite, 0 = off) --schemas Specify the database schemas to introspect --local-d1 Generate a Prisma schema from a local Cloudflare D1 database

Examples

Basic introspection

prisma db pull

Preview without writing

prisma db pull --print

Outputs schema to terminal for review.

Force overwrite

prisma db pull --force

Replaces schema file, losing any manual customizations.

Workflow

Starting from existing database

Initialize Prisma:

prisma init

Configure database URL

Pull schema:

prisma db pull

Review and customize generated schema

Generate client:

prisma generate

Syncing changes from database

When database changes are made outside Prisma:

prisma db pull
prisma generate

Generated Schema Example

Database tables become Prisma models:

-- Database tables
CREATE TABLE users (
 id SERIAL PRIMARY KEY,
 email VARCHAR(255) UNIQUE NOT NULL,
 name VARCHAR(100)
);

CREATE TABLE posts (
 id SERIAL PRIMARY KEY,
 title VARCHAR(255) NOT NULL,
 author_id INTEGER REFERENCES users(id)
);

Becomes:

model users {
 id Int @id @default(autoincrement())
 email String @unique @db.VarChar(255)
 name String? @db.VarChar(100)
 posts posts[]
}

model posts {
 id Int @id @default(autoincrement())
 title String @db.VarChar(255)
 author_id Int?
 users users? @relation(fields: [author_id], references: [id])
}

Post-Introspection Cleanup

After db pull, consider:

Rename models to PascalCase:

model User { // Was: users
 @@map("users")
}

Rename fields to camelCase:

authorId Int? @map("author_id")

Add relation names for clarity:

author User? @relation("PostAuthor", fields: [authorId], references: [id])

Add documentation:

/// User account information
model User {
 /// Primary email for authentication
 email String @unique
}

MongoDB Introspection

For MongoDB, db pull samples documents to infer schema:

prisma db pull

May require manual refinement since MongoDB is schemaless.

Warning

db pull overwrites your schema file. Always:

• Commit current schema before pulling

• Use --print to preview first

• Backup customizations you want to keep