What problem does it solve? Designing a PostgreSQL schema without deep platform knowledge leads to common mistakes: missing foreign key indexes, wrong data types like timestamp without time zone, silent performance traps, and painful schema migrations later. This Skill encodes PostgreSQL-specific best practices so schemas are correct and performant from the start. ## Core Features & Use Cases - Data Type Guidance: Recommends correct types for every column, including TIMESTAMPTZ for time, NUMERIC for money, TEXT for strings, identity columns over serial, and JSONB with GIN indexing for semi-structured data. - Indexing & Constraints: Covers B-tree, composite, partial, expression, covering, GIN, GiST, and BRIN indexes, plus primary keys, foreign keys with manual indexing, UNIQUE NULLS NOT DISTINCT, CHECK, and EXCLUDE constraints. - Advanced Patterns: Addresses partitioning strategies, row-level security, generated columns, upsert-friendly design, update-heavy and insert-heavy workload tuning, safe schema evolution, and extensions like TimescaleDB, PostGIS, and pgvector. - Use Case: When designing a new orders table, apply this Skill to get an identity primary key, a foreign key with its own index, a CHECK constraint on status values, NUMERIC for totals, and TIMESTAMPTZ defaults, avoiding the gotchas that typically surface only in production. ## Quick Start Ask the AI to design or review a PostgreSQL table schema for your use case, such as "design a PostgreSQL schema for an orders and users system following best practices."
