---

name: database-schema-designer description: 'Design PostgreSQL, MongoDB, or Redis schemas with entities, relationships, indexes, constraints, migrations, query optimization, and backward-compatible schema evolution'

Purpose

Design efficient, scalable database schemas optimized for application access patterns with proper indexing and migration strategies.

When to Use

Use this skill when the task involves:

• Designing new database tables, collections, or key structures
• Creating or modifying indexes for query performance
• Writing database migrations (Prisma, TypeORM, Drizzle, raw SQL)
• Optimizing slow queries or analyzing execution plans
• Planning schema changes with backward compatibility

Constraints

• Always consider query patterns first before designing schema
• Include migration rollback scripts for every schema change
• Apply least-privilege for database roles and permissions
• Use explicit constraints (NOT NULL, UNIQUE, CHECK) at database level
• Prefer UUID or ULID over auto-increment for distributed systems
• Document breaking changes and data migration requirements

Execution Steps

1. Identify entities — List domain entities and their attributes
2. Map relationships — Define 1:1, 1:N, M:N relationships with cardinality
3. Design schema — Create tables/collections with appropriate types
4. Add indexes — Create indexes based on WHERE, JOIN, and ORDER BY patterns
5. Write migrations — Generate up/down migration scripts
6. Validate performance — Check EXPLAIN plans for critical queries
7. Document changes — Note breaking changes and required data backfills

Examples

Input

Design a schema for a blog with posts, authors, and tags (many-to-many)

Output

Copy-- PostgreSQL schema
CREATE TABLE authors (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  email VARCHAR(255) NOT NULL UNIQUE,
  name VARCHAR(255) NOT NULL,
  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
);

CREATE TABLE posts (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  author_id UUID NOT NULL REFERENCES authors(id) ON DELETE CASCADE,
  title VARCHAR(500) NOT NULL,
  slug VARCHAR(500) NOT NULL UNIQUE,
  content TEXT,
  published_at TIMESTAMPTZ,
  created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
);

CREATE TABLE tags (
  id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
  name VARCHAR(100) NOT NULL UNIQUE
);

CREATE TABLE post_tags (
  post_id UUID REFERENCES posts(id) ON DELETE CASCADE,
  tag_id UUID REFERENCES tags(id) ON DELETE CASCADE,
  PRIMARY KEY (post_id, tag_id)
);

-- Indexes for common queries
CREATE INDEX idx_posts_author ON posts(author_id);
CREATE INDEX idx_posts_published ON posts(published_at) WHERE published_at IS NOT NULL;
CREATE INDEX idx_posts_slug ON posts(slug);

Related Skills

• nest-backend-service-builder — Implement repository layer
• system-planning-assistant — High-level data architecture decisions