Database Migration Skill

Expert in PostgreSQL schema management and migrations following project conventions.

When to Use This Skill

Use this skill when:

• •Creating new tables
• •Adding or modifying columns
• •Creating indexes or constraints
• •Any database schema change
• •Migration-related tasks

CRITICAL: Migration-First Workflow (MANDATORY)

ALL database changes MUST have a migration file.

Without a migration, changes will NOT deploy to production.

Migration Workflow

1. •

   Create Migration File

   bash

   # In docker/postgres/migrations/
   # Name: NNN_description.sql (e.g., 003_add_user_roles.sql)
   

2. •

   Write Idempotent Migration

   sql

   BEGIN;
   
   -- Create table only if it doesn't exist
   CREATE TABLE IF NOT EXISTS users (
     id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
     "firstName" TEXT NOT NULL,
     "lastName" TEXT NOT NULL,
     email TEXT UNIQUE NOT NULL,
     "createdAt" TIMESTAMP DEFAULT NOW()
   );
   
   -- Add column only if it doesn't exist
   DO $$
   BEGIN
     IF NOT EXISTS (
       SELECT 1 FROM information_schema.columns
       WHERE table_name = 'users' AND column_name = 'role'
     ) THEN
       ALTER TABLE users ADD COLUMN role TEXT DEFAULT 'user';
     END IF;
   END $$;
   
   -- Record migration
   INSERT INTO schema_migrations (version, description)
   VALUES ('003', 'Add user roles')
   ON CONFLICT (version) DO NOTHING;
   
   COMMIT;
   

3. •

   Update init.sql

   bash

   # Add the same schema to docker/postgres/init.sql
   # For fresh database installations
   

4. •

   Test Migration Locally

   bash

   # Apply migration to test database
   docker exec -i st44-db-test psql -U postgres -d st44_test < docker/postgres/migrations/003_add_user_roles.sql
   
   # Verify it worked
   docker exec -it st44-db-test psql -U postgres -d st44_test -c "\d users"
   

5. •

   Test Idempotency

   bash

   # Run migration again - should not error
   docker exec -i st44-db-test psql -U postgres -d st44_test < docker/postgres/migrations/003_add_user_roles.sql
   

CRITICAL: camelCase Column Names

ALL columns MUST use camelCase with double quotes.

✅ CORRECT

CREATE TABLE users (
  id UUID PRIMARY KEY,
  "firstName" TEXT NOT NULL,
  "lastName" TEXT NOT NULL,
  "createdAt" TIMESTAMP DEFAULT NOW()
);

❌ FORBIDDEN

CREATE TABLE users (
  id UUID PRIMARY KEY,
  first_name TEXT NOT NULL,  -- NO snake_case!
  last_name TEXT NOT NULL,   -- NO snake_case!
);

Why: Consistency across entire stack (TypeScript, API, database).

Migration File Requirements

File Naming

docker/postgres/migrations/
├── 001_initial_schema.sql
├── 002_add_households.sql
├── 003_add_user_roles.sql
└── 004_add_tasks_table.sql

• •Format: NNN_description.sql
• •Number: Sequential (001, 002, 003...)
• •Description: Lowercase with underscores

File Structure

BEGIN;

-- Your schema changes here
-- Use IF NOT EXISTS for idempotency

-- Always record the migration
INSERT INTO schema_migrations (version, description)
VALUES ('NNN', 'Description of changes')
ON CONFLICT (version) DO NOTHING;

COMMIT;

Idempotency Patterns

Creating Tables

CREATE TABLE IF NOT EXISTS table_name (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  "columnName" TEXT NOT NULL
);

Adding Columns

DO $$
BEGIN
  IF NOT EXISTS (
    SELECT 1 FROM information_schema.columns
    WHERE table_name = 'users' AND column_name = 'role'
  ) THEN
    ALTER TABLE users ADD COLUMN role TEXT DEFAULT 'user';
  END IF;
END $$;

Creating Indexes

CREATE INDEX IF NOT EXISTS idx_users_email ON users(email);

Adding Constraints

DO $$
BEGIN
  IF NOT EXISTS (
    SELECT 1 FROM pg_constraint
    WHERE conname = 'users_email_unique'
  ) THEN
    ALTER TABLE users ADD CONSTRAINT users_email_unique UNIQUE (email);
  END IF;
END $$;

Testing Checklist

Before marking migration complete:

• • Migration file created in docker/postgres/migrations/NNN_description.sql
• • init.sql updated with same schema
• • Migration uses BEGIN/COMMIT transaction
• • All operations are idempotent (IF NOT EXISTS)
• • camelCase column names with double quotes
• • schema_migrations table updated
• • Tested on local database
• • Tested idempotency (ran twice without errors)
• • Verified with \d table_name in psql

Common Database Operations

UUID Primary Keys

CREATE TABLE table_name (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid()
);

Timestamps

"createdAt" TIMESTAMP DEFAULT NOW(),
"updatedAt" TIMESTAMP DEFAULT NOW()

Foreign Keys

"userId" UUID NOT NULL REFERENCES users(id) ON DELETE CASCADE

Enums (Use TEXT with CHECK)

role TEXT CHECK (role IN ('admin', 'parent', 'child'))

Testing Migrations

Start Test Database

npm run db:test:up

Apply Migration

docker exec -i st44-db-test psql -U postgres -d st44_test < docker/postgres/migrations/003_add_user_roles.sql

Verify Schema

docker exec -it st44-db-test psql -U postgres -d st44_test -c "\d users"

Test Idempotency

# Run same migration again - should not error
docker exec -i st44-db-test psql -U postgres -d st44_test < docker/postgres/migrations/003_add_user_roles.sql

Stop Test Database

npm run db:test:down

Workflow

1. •Read the optimized agent spec: .claude/agents/agent-database.md
2. •Understand what schema changes are needed
3. •Create migration file with next sequential number
4. •Write idempotent SQL with camelCase columns
5. •Update init.sql with same schema
6. •Test locally on test database
7. •Verify idempotency (run twice)
8. •Only then commit and push

Reference Files

For detailed patterns and examples:

• •.claude/agents/agent-database.md - Complete agent specification
• •docker/postgres/init.sql - Current database schema
• •docker/postgres/migrations/ - Existing migration examples
• •CLAUDE.md - Project-wide conventions

Deployment Guarantee

If you create a migration file and it passes testing: ✅ It WILL deploy to production ✅ It WILL run automatically ✅ Schema changes are guaranteed

If you DON'T create a migration file: ❌ Changes will NOT deploy ❌ Production will be out of sync ❌ Backend will fail with schema errors

This is why migration-first is mandatory.

Success Criteria

Before marking work complete:

• • Migration file exists and is numbered correctly
• • init.sql updated
• • All columns use camelCase
• • Migration is idempotent
• • Tested locally and verified
• • Idempotency tested (ran twice)
• • No errors when applying migration