epic-database_skill

Overview

This skill is a focused guide for using Prisma with SQLite and LiteFS in the Epic Stack. It explains schema design, migrations, multi-region considerations, and pragmatic query optimization. The goal is practical, production-ready advice for building fast, maintainable apps with minimal complexity.

How this skill works

The skill walks through Prisma schema patterns (models, relations, indexes), migration workflows, and seed scripts. It covers SQLite-specific operation with LiteFS for multi-region setups, including how to detect the primary instance for writes and how reads behave on replicas. It also shows pragmatic query examples, transactions, and logging to measure and optimize performance.

When to use it

• Designing or refactoring a Prisma schema for an Epic Stack app
• Creating and applying migrations locally and in production
• Running a multi-region SQLite deployment with LiteFS where primary/replica behavior matters
• Writing seed scripts, transactional flows, or complex queries with pagination
• Diagnosing slow queries and deciding where to add indexes

Best practices

• Fetch only the fields you need using select; avoid over-fetching entire models
• Prefer simple, readable queries and optimize only after measuring real bottlenecks
• Index foreign keys and fields used frequently in where/orderBy; add composite indexes for measured needs
• Follow a widen-then-narrow migration strategy for zero-downtime schema changes
• Use transactions for multi-step, related writes and ensure writes run on the LiteFS primary

Example use cases

• Create a User and related Note models with proper indexes and cascade delete
• Implement paginated post listing with selective relation includes and total count
• Run seed scripts to populate roles, permissions, and example users before first deploy
• Detect LiteFS primary before performing write operations and redirect reads to replicas as needed
• Log slow Prisma queries and add indexes based on observed durations