Cursor Rules

October 17, 2025 · View on GitHub

Overview

This page explains the purpose of Cursor Rules for this project. Use it as a single place to capture and evolve the rules that guide how Cursor should operate (architecture constraints, generation workflows, naming, and any process conventions).

What to add here in the future:

  • High-level rules that Cursor must follow (brief and actionable)
  • Pointers to detailed docs elsewhere (architecture, Hygen, setup, etc.)
  • Any project-specific conventions that affect Cursor usage

Keep this document concise; link out to details instead of duplicating them.

Current Rules (brief)

  • Follow docs/ as the single source of truth.
  • Enforce hexagonal boundaries: services ↔ domain; repositories ↔ persistence; use mappers for conversions.
  • Type safety first: no any in public APIs; explicit types.
  • Naming: DTOs/domain/controllers use camelCase; persistence (entities) use snake_case; map between them.
  • Hygen: prefer JSON-driven runs with DATA_FILE; generate entities via .hygen-entities-generator/entities-generator.json.
  • Custom schema: SQL at .hygen/generate-migration/sql-script.sql.
  • Command execution from chat: use bash bash -lc "source ~/.bashrc && npm run <script> -- <args>".
  • Long/compound commands: first show options (e.g., scripts-index), then execute on user confirmation.
  • After code level changes specially Hygen generation: run lint and build.
  • Also added developer checklist and review process sections

Previous: Documentation Standards

Next: Cursor Commands