create-database

Designs PostgreSQL schemas and architecture documentation from domain source materials.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill create-database-gozonerd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: create-database
Source: https://github.com/gozonerd/shadow-ai-assessment/tree/main/.claude/skills/create-database
Command: npx skills add https://github.com/gozonerd/shadow-ai-assessment --skill create-database-gozonerd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Designing a relational database schema from scratch requires translating domain documents into normalized tables, ENUM types, constraints, and indexes, which is error-prone and inconsistent without a structured protocol. ## Core Features & Use Cases - Schema Authoring: Produces a transaction-wrapped db/schema.sql with ENUM types, tables, constraints, indexes, views, and COMMENT ON annotations. - Architecture Documentation: Generates a 7-section docs/architecture.md covering design rationale, ER overview, deployment notes, and migration strategy. - Engine Selection Guidance: Includes a decision tree for choosing PostgreSQL versus SQLite or document stores based on access patterns and data shape. - Use Case: Given markdown files describing a taxonomy domain, walk through entity extraction, ER modeling, and produce a complete 11-table PostgreSQL schema with gap-analysis views ready for deployment. ## Quick Start Ask the AI to design a database schema for your project by pointing it at your domain source documents and target repository.

Frequently Asked Questions about create-database

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design a PostgreSQL schema from domain documents?

Extract entities, relationships, and controlled vocabularies from your source materials, then confirm an ER model before writing DDL. This skill walks through that protocol and outputs db/schema.sql plus docs/architecture.md.

Should I use PostgreSQL or SQLite for my project database?

Choose PostgreSQL for multi-user access, native ENUM types, and complex JOINs across many tables. SQLite fits single-user, sub-100K-row scenarios with no server; the skill includes a decision tree for this choice.

Should database table names be singular or plural?

This skill mandates singular nouns (taxonomy, not taxonomies) following DBA-traditional convention from Joe Celko and Microsoft. Singular avoids irregular-plural debates like octopi versus octopuses and reads naturally in JOIN clauses.

Why do PostgreSQL foreign keys need explicit indexes?

PostgreSQL does not automatically index foreign key columns, so unindexed FKs turn JOINs into sequential scans. The skill requires a CREATE INDEX for every FK column with descriptive names like idx_table_column.

When should I not use this database design skill?

Skip it when the database already exists and only needs migrations, when the project needs a document or graph store, or when modifying an existing schema. It is relational-first and PostgreSQL-first for greenfield design.