Why does this skill prefer BIGINT over SERIAL?

The skill follows modern SQL standards by recommending 'GENERATED ALWAYS AS IDENTITY', which is more robust and standard-compliant than the legacy PostgreSQL SERIAL type.

Can this skill help with AI-specific database features?

Yes, it includes guidance on using 'pgvector' for vector similarity searches, which is essential for managing AI embeddings within PostgreSQL.

How does it handle performance for large datasets?

It provides specific guidance on declarative partitioning (Range, List, Hash), BRIN indexing for time-series data, and partitioning strategies for tables exceeding 100 million rows.

Does it support data integrity via constraints?

Absolutely. It covers Primary Keys, Foreign Keys with proper indexing, Unique constraints with NULL handling, and Check/Exclude constraints for complex business logic.

PostgreSQL Table Design

Name: PostgreSQL Table Design
Author: arthur0824hao

byarthur0824hao

データベース管理

Designs optimized, production-grade PostgreSQL schemas following industry best practices and performance patterns.

概要

The PostgreSQL Table Design skill provides Claude with a comprehensive framework for architecting robust relational databases. It enforces strict data integrity through normalization, proper constraint usage, and precise data type selection—such as preferring TIMESTAMPTZ over TIMESTAMP and IDENTITY columns over SERIAL. By incorporating advanced concepts like GIN/GiST indexing, Row-Level Security (RLS), and partitioning strategies, this skill ensures that database schemas are not only performant under heavy workloads but also scalable and easy to maintain.

主な機能

Schema normalization (3NF) and strategic denormalization patterns
Advanced performance tuning for update-heavy and insert-heavy workloads
Comprehensive indexing strategies including B-tree, GIN, GiST, and partial indexes
Implementation of Row-Level Security and complex constraints
Expert data type selection for performance and precision (JSONB, Vector, UUID)
0 GitHub stars

ユースケース

Designing specialized storage for time-series data or vector embeddings
Architecting a new database schema from scratch with scalable foundations
Refactoring existing SQL tables to resolve performance bottlenecks and bloat

概要

主な機能

Schema normalization (3NF) and strategic denormalization patterns
Advanced performance tuning for update-heavy and insert-heavy workloads
Comprehensive indexing strategies including B-tree, GIN, GiST, and partial indexes
Implementation of Row-Level Security and complex constraints
Expert data type selection for performance and precision (JSONB, Vector, UUID)
0 GitHub stars

ユースケース

Designing specialized storage for time-series data or vector embeddings
Architecting a new database schema from scratch with scalable foundations
Refactoring existing SQL tables to resolve performance bottlenecks and bloat