What is drizzle-schema-guide?

This Claude skill streamlines the management of Drizzle ORM database schemas, enabling developers to safely add tables, modify columns, and handle migrations. It integrates best practices for TypeScript and Neon Serverless Postgres to ensure data integrity and efficient database evolution while providing automated guidance for schema synchronization.

When should I use drizzle-schema-guide?

drizzle-schema-guide is useful in the following scenarios: • Expanding database architecture by adding new tables with standardized UUID primary keys and audit timestamps. • Modifying existing schemas by adding columns with NOT NULL constraints and safe defaults to prevent breaking changes in production. • Orchestrating database migrations by distinguishing between development (db:push) and production (db:generate/migrate) workflows. • Automating the export of schema types to maintain end-to-end type safety across the TypeScript application. • Safely inspecting and debugging database structures using integrated Drizzle Studio commands.

name	drizzle-schema-guide
description	Modify Drizzle ORM database schema safely. Use when user mentions "add table", "new column", "database schema", "migration", or "drizzle".
allowed-tools	Read, Grep, Glob

Drizzle Schema Modifications

This project uses Drizzle ORM with Neon Serverless Postgres.

Project Files

Schema: db/schema.ts
Config: drizzle.config.ts
Connection: db/index.ts

Current Schema

// x_account_cache - Caches X account analysis
export const xAccountCache = pgTable('x_account_cache', {
  id: uuid('id').primaryKey().defaultRandom(),
  xHandle: text('x_handle').unique().notNull(),
  searchResponse: jsonb('search_response').notNull(),
  createdAt: timestamp('created_at', { withTimezone: true }).defaultNow(),
  expiresAt: timestamp('expires_at', { withTimezone: true }).notNull(),
});

// usage_tracking - Tracks premium image quota
export const usageTracking = pgTable('usage_tracking', {
  id: uuid('id').primaryKey().defaultRandom(),
  userIdentifier: text('user_identifier').unique().notNull(),
  premiumImagesCount: integer('premium_images_count').default(0),
  lastResetAt: timestamp('last_reset_at', { withTimezone: true }).defaultNow(),
  createdAt: timestamp('created_at', { withTimezone: true }).defaultNow(),
  updatedAt: timestamp('updated_at', { withTimezone: true }).defaultNow(),
});

Instructions

Adding a new table:

import { pgTable, uuid, text, timestamp, integer, index } from 'drizzle-orm/pg-core';

export const newTable = pgTable('new_table', {
  id: uuid('id').primaryKey().defaultRandom(),
  // columns here
  createdAt: timestamp('created_at', { withTimezone: true }).defaultNow(),
}, (table) => [
  index('idx_name').on(table.column),  // optional indexes
]);

// Type exports
export type NewTable = typeof newTable.$inferSelect;
export type InsertNewTable = typeof newTable.$inferInsert;

Update db/index.ts to export the new table:

export { db } from './index';
export { xAccountCache, usageTracking, newTable } from './schema';

Apply schema changes:
- Development: npm run db:push (direct push, no migration files)
- Production: npm run db:generate then npm run db:migrate
Inspect database: npm run db:studio opens Drizzle Studio

Common Column Types

uuid('id').primaryKey().defaultRandom()
text('name').notNull()
text('email').unique()
integer('count').default(0)
boolean('active').default(true)
jsonb('data').notNull()
timestamp('created_at', { withTimezone: true }).defaultNow()

Examples

"Add a feedback table" → Add table definition to db/schema.ts, export in db/index.ts
"Add email column to usage" → Modify usageTracking in schema

Guardrails

READ-ONLY by default - review changes before applying
Never run db:push against production
Always generate migrations for production changes
Backup data before destructive migrations
Confirm with user before running any db commands
Add NOT NULL columns with defaults to avoid breaking existing rows

drizzle-schema-guide

When & Why to Use This Skill

Use Cases