$TLOGZ Prompt Flow
~/development/database-schema-sql-builder.md
$catdatabase-schema-sql-builder.md

Database Schema & SQL Builder

Design normalized database schemas, write optimized SQL queries, and generate migration scripts.

Best
gpt-4o
Good
claude-sonnet-4, gemini-2.5-pro
Limited
claude-haiku, gpt-4o-mini
Updated
2026-05-22
workflow

You are a database architect. Design a production-ready database schema.

Use Case: {{useCase}} Core Entities: {{entities}} Database Type: {{databaseType || "PostgreSQL"}} Key Queries: {{queries || "CRUD operations, search, listing"}} Performance Requirements: {{performance || "Standard workload"}}

Schema Design

1. Entity Relationship Overview

[Entity] 1---* [Entity] : [relationship description] [Entity] *---* [Entity] : [through join table] ...

2. Table Definitions

Table: [entity_name]

ColumnTypeConstraintsDefaultIndexNotes
idUUIDPK, NOT NULLgen_random_uuid()PKPrimary key
[field][type][constraints][default][index type][notes]
created_atTIMESTAMPTZNOT NULLNOW()
updated_atTIMESTAMPTZNOT NULLNOW()
sql
CREATE TABLE [entity_name] ( id UUID PRIMARY KEY DEFAULT gen_random_uuid(), ... ); CREATE INDEX idx_[entity]_[field] ON [entity_name]([field]);

3. Relationships

ParentChildTypeForeign KeyON DELETE
[table][table]1:N[column]CASCADE/SET NULL/RESTRICT
[table][table]N:M[join table]CASCADE

4. Migration Scripts

Migration 001: Initial Schema

sql
-- Up [CREATE statements] -- Down [DROP statements]

Migration 002: [Description]

sql
-- Up [ALTER statements] -- Down [ROLLBACK statements]

5. Optimized Queries

List [Entity] with Pagination

sql
SELECT * FROM [entity] WHERE [conditions] ORDER BY [field] [ASC/DESC] LIMIT $1 OFFSET $2;

Search [Entity]

sql
SELECT * FROM [entity] WHERE [text_field] ILIKE '%' || $1 || '%' OR [text_field] ILIKE '%' || $1 || '%' ORDER BY [relevance_field] DESC; -- Add full-text search index for production: -- CREATE INDEX idx_[entity]_fts ON [entity] USING GIN(to_tsvector('english', [field]));

Aggregation Query (Dashboard)

sql
SELECT DATE_TRUNC('[period]', created_at) AS period, COUNT(*) AS total, COUNT(DISTINCT [column]) AS unique_count, AVG([numeric_field]) AS average FROM [entity] WHERE created_at > NOW() - INTERVAL '[interval]' GROUP BY period ORDER BY period DESC;

6. Performance Notes

  • Expected Data Size: [rows/month, total size]
  • Index Strategy: Which columns need B-tree, GIN, or partial indexes
  • Query Patterns: Read-heavy, write-heavy, or mixed
  • Partitioning Recommendation: If applicable
  • Connection Pooling: Recommended pool size

7. Normalization Level

  • Current: [3NF, Star Schema, etc.]
  • Denormalization Candidates: Fields that could be cached for performance

Output with code blocks for all SQL, | table | for column definitions, bold for table names, and --- for migration separators.

variables
^Enter
guide
how to use
  • Open the Database Schema & SQL Builder workflow in your AI chat interface.
  • Replace the variables in [brackets] with your specific inputs.
  • For best results, use gpt-4o as the target model.
  • Review the generated output and iterate by refining your inputs.
  • Save your final result and share it with your team.
best use cases
  • Quickly generate development-specific content with structured prompts.
  • Standardize development workflows across your team using a shared template.
  • Onboard new team members with a repeatable development process.
  • Automate sql tasks with AI-powered development workflows.
  • Automate database tasks with AI-powered development workflows.
  • Automate schema-design tasks with AI-powered development workflows.
examples
  • Use Database Schema & SQL Builder to create a sql project from scratch.
  • Adapt Database Schema & SQL Builder for a different development domain with custom variables.
  • Combine Database Schema & SQL Builder with other workflows in the development category for a complete pipeline.
  • Run Database Schema & SQL Builder with multiple AI models to compare output quality.
  • Schedule Database Schema & SQL Builder as a recurring development task.
variations
  • Simplified version: remove optional variables for faster results.
  • Advanced version: add custom validation steps after generation.
  • Batch version: run Database Schema & SQL Builder on multiple inputs sequentially.
  • sql-focused variant: emphasize sql best practices in the prompt.
  • database-focused variant: emphasize database best practices in the prompt.
common mistakes
  • Skipping variable customization — always replace [bracketed] placeholders.
  • Using the wrong AI model tier for complex outputs.
  • Not iterating on the first result — refinement improves quality significantly.
  • Ignoring sql best practices when customizing the prompt.
  • Using gpt-4o outside its optimal use case for this workflow.
Remix