What is the recommended way to handle strings and lengths?

The skill recommends using the TEXT type over VARCHAR(n). If length limits are required, it suggests using a CHECK constraint instead of a hard-coded character limit.

What primary key type does this skill recommend?

The skill prefers BIGINT GENERATED ALWAYS AS IDENTITY for most tables, suggesting UUID only when global uniqueness or opacity is required.

When is partitioning recommended?

Partitioning is suggested for tables exceeding 100 million rows or where specific maintenance tasks like periodic data pruning are required.

How does it handle indexing for foreign keys?

Since PostgreSQL does not auto-index FK columns, this skill mandates manual index creation on referencing columns to speed up joins and prevent locking issues.

PostgreSQL Table Design

Name: PostgreSQL Table Design
Author: timescale

bytimescale

•

170

•

Database Management

Optimizes PostgreSQL database schemas using industry-standard patterns for data types, indexing, and high-performance architecture.

This skill acts as a specialized architectural guide for designing robust and performant PostgreSQL databases within AI coding workflows. It provides domain-specific knowledge to help developers implement best practices such as selecting optimal data types (e.g., BIGINT over SERIAL), enforcing normalization standards, and identifying the correct indexing strategies for complex query patterns. By codifying rules for Row-Level Security, partitioning, and safe schema evolution, it ensures that generated SQL code is production-ready, scalable, and free from common pitfalls like unindexed foreign keys or inefficient storage strategies.

Key Features

01170 GitHub stars

02Performance tuning for update-heavy and insert-heavy workloads

03Automated indexing patterns for foreign keys and selective filters

04Comprehensive guidance for declarative partitioning and Row-Level Security

05Optimized data type selection including BIGINT Identity and TIMESTAMPTZ

06Strict adherence to PostgreSQL naming conventions and snake_case identifiers

Use Cases

01Designing a high-performance relational schema for a scalable SaaS backend

02Implementing complex multi-tenant access controls using Row-Level Security policies

03Refactoring legacy database tables to reduce bloat and improve query speed

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add timescale/pg-aiguide design-postgres-tables

For use in Claude.ai and ChatGPT

Download Skill