When should I use TIMESTAMPTZ instead of TIMESTAMP?

The skill strongly recommends TIMESTAMPTZ for all event data to ensure timezone consistency across distributed systems, avoiding the common pitfalls of wall-clock time variations.

Why does this skill recommend BIGINT over SERIAL?

Modern PostgreSQL standards prefer 'GENERATED ALWAYS AS IDENTITY' (which uses BIGINT) because it is SQL-compliant, handles sequence permissions more effectively, and avoids the legacy pitfalls of the SERIAL pseudo-type.

Does this skill help with indexing strategy?

Yes, it provides specific rules for when to use B-tree, GIN (for JSONB/Arrays), GiST (for ranges/geometry), and BRIN (for large time-series data), as well as guidance on composite and partial indexes.

Can it help with large-scale data partitioning?

Absolutely. It covers Range, List, and Hash partitioning strategies, specifically recommending them for tables exceeding 100 million rows to maintain query performance.

PostgreSQL Table Design

Name: PostgreSQL Table Design
Author: 3commas-io

by3commas-io

Database Management

Architects optimized PostgreSQL schemas using industry best practices for performance, scalability, and data integrity.

About

The PostgreSQL Table Design skill provides comprehensive architectural guidance for building production-grade database schemas. It assists developers in selecting the most efficient data types, implementing advanced indexing strategies like GIN and GiST, and configuring partitioning for large-scale datasets. By incorporating best practices for normalization, row-level security, and workload-specific optimizations, this skill ensures that database structures are resilient, performant, and maintainable within the Claude Code environment.

Key Features

Declarative partitioning patterns for Range, List, and Hash
Advanced indexing strategies for B-tree, GIN, GiST, and BRIN
Workload-specific optimizations for insert-heavy and update-heavy tables
0 GitHub stars
Expert data type selection including TIMESTAMPTZ, BIGINT, and JSONB
Implementation of safe schema evolution and transactional DDL

Use Cases

Designing high-performance schemas for new SaaS applications
Implementing complex constraints and Row-Level Security for multi-tenant apps
Refactoring legacy tables to improve query speed and reduce storage bloat

About

Key Features

Declarative partitioning patterns for Range, List, and Hash
Advanced indexing strategies for B-tree, GIN, GiST, and BRIN
Workload-specific optimizations for insert-heavy and update-heavy tables
0 GitHub stars
Expert data type selection including TIMESTAMPTZ, BIGINT, and JSONB
Implementation of safe schema evolution and transactional DDL

Use Cases

Designing high-performance schemas for new SaaS applications
Implementing complex constraints and Row-Level Security for multi-tenant apps
Refactoring legacy tables to improve query speed and reduce storage bloat