AI-Friendly Backend Template

A sample backend template that leverages AI tooling to create a fully typed, auto-documented API server.

Features

• Type Safety: End-to-end type safety from database to API endpoints using Prisma and TypeBox
• Auto-Documentation: Automatic Swagger documentation generation for API exploration and testing
• AI-Optimized: Structured for efficient AI-assisted development using tools like Cursor
• Production Ready: Includes deployment configuration for Railway and database integration with Supabase
• Developer Experience: Streamlined workflow for rapid iteration and schema changes

Tech Stack

• Database: Supabase (PostgreSQL)
• ORM: Prisma
• Backend Framework: Elysia
• Schema Validation: TypeBox
• Deployment: Railway
• Development: Cursor (AI-assisted IDE)

Project Structure

• Dockerfile - Container configuration for building

• prisma/

  • schema.prisma - Database schema definition

• src/

  • endpoints/ - API endpoint handlers
    • ... - Individual endpoint files
  • server.ts - Main Alesia application
  • types.ts - Typebox schemas and type definitions
  • db.ts - Database operations and queries
  • p_db.md - Database-related prompt
  • p_endpoints.md - Endpoint-related prompt
  • p_types.md - Type-related prompt

How It Works

This repository is designed to make AI-assisted backend development smooth and reliable. Here's how it works:

1. Database Schema as Source of Truth

The database schema (in prisma/schema.prisma) serves as the single source of truth. After making changes to your database in Supabase:

1. Run bunx prisma db pull && bunx prisma generate to update your local schema
2. The Prisma schema provides the foundation for generating the rest of your backend

2. Prompts as Documentation

The src/p_*.md files serve as both documentation and AI prompts:

• p_types.md - Templates for generating TypeBox schemas
• p_db.md - Templates for database operations
• p_endpoints.md - Templates for API endpoints

These prompts ensure consistent code generation and serve as documentation for the codebase patterns.

3. AI-Assisted Code Generation

Using Cursor (or similar AI-enabled IDE):

1. Select the file you want to update
2. Use CMD+K to open the inline editor
3. Reference relevant schema and prompt files
4. Describe your changes
5. Let the AI generate the appropriate code

For example, after adding a new table: