Backend Development Guidelines
SkillDatabases & databackend claude skill that gives an AI agent a set of backend development guidelines for Node.js, Express, and TypeScript microservices. When the agent writes routes, controllers, services, repositories, or middleware, it follows the same house rules instead of inventing its own patterns.
Available today. Use it from your connected AI after setup.
No other account needed.
Have an agent that supports loading skills.
Then ask your AI: use the Backend Development Guidelines skill
What your AI can do with it
- Guides creation of routes, controllers, services, repositories, and middleware
- Covers Express API work in Node.js and TypeScript microservices
- Directs Prisma database access patterns
- Applies Zod validation and Sentry error tracking conventions
- Enforces unifiedConfig and dependency injection practices
- Handles async code according to the guidelines
Getting started
- Have an agent that supports loading skills.
- Add the backend-dev-guidelines skill to the agent's available skills.
- Ask the agent to write or modify backend code; it loads the skill when creating routes, controllers, services, repositories, or middleware, or when working with Express APIs, Prisma, Sentry, Zod, unif
What this skill tells your AI
The instructions your AI receives, as published by davila7/claude-code-templates in cli-tool/components/skills/development/backend-dev-guidelines/SKILL.md and read by ahel’s review.
Purpose
Establish consistency and best practices across backend microservices (blog-api, auth-service, notifications-service) using modern Node.js/Express/TypeScript patterns.
When to Use This Skill
Automatically activates when working on:
- Creating or modifying routes, endpoints, APIs
- Building controllers, services, repositories
- Implementing middleware (auth, validation, error handling)
- Database operations with Prisma
- Error tracking with Sentry
- Input validation with Zod
- Configuration management
- Backend testing and refactoring
Quick Start
New Backend Feature Checklist
- Route: Clean definition, delegate to controller
- Controller: Extend BaseController
- Service: Business logic with DI
- Repository: Database access (if complex)
- Validation: Zod schema
- Sentry: Error tracking
- Tests: Unit + integration tests
- Config: Use unifiedConfig
New Microservice Checklist
- Directory structure (see architecture-overview.md)
- instrument.ts for Sentry
- unifiedConfig setup
- BaseController class
- Middleware stack
- Error boundary
- Testing framework
Architecture Overview
Layered Architecture
HTTP Request
↓
Routes (routing only)
↓
Controllers (request handling)
↓
Services (business logic)
↓
Repositories (data access)
↓
Database (Prisma)
Key Principle: Each layer has ONE responsibility.
See architecture-overview.md for complete details.
Directory Structure
service/src/
├── config/ # UnifiedConfig
├── controllers/ # Request handlers
├── services/ # Business logic
├── repositories/ # Data access
├── routes/ # Route definitions
├── middleware/ # Express middleware
├── types/ # TypeScript types
├── validators/ # Zod schemas
├── utils/ # Utilities
├── tests/ # Tests
├── instrument.ts # Sentry (FIRST IMPORT)
├── app.ts # Express setup
└── server.ts # HTTP server
Naming Conventions:
- Controllers:
PascalCase-UserController.ts - Services:
camelCase-userService.ts - Routes:
camelCase + Routes-userRoutes.ts - Repositories:
PascalCase + Repository-UserRepository.ts
Core Principles (7 Key Rules)
1. Routes Only Route, Controllers Control
// ❌ NEVER: Business logic in routes
router.post('/submit', async (req, res) => {
// 200 lines of logic
});
// ✅ ALWAYS: Delegate to controller
router.post('/submit', (req, res) => controller.submit(req, res));
2. All Controllers Extend BaseController
export class UserController extends BaseController {
async getUser(req: Request, res: Response): Promise<void> {
try {
const user = await this.userService.findById(req.params.id);
this.handleSuccess(res, user);
} catch (error) {
this.handleError(error, res, 'getUser');
}
}
}
3. All Errors to Sentry
try {
await operation();
} catch (error) {
Sentry.captureException(error);
throw error;
}
4. Use unifiedConfig, NEVER process.env
// ❌ NEVER
const timeout = process.env.TIMEOUT_MS;
// ✅ ALWAYS
import { config } from './config/unifiedConfig';
const timeout = config.timeouts.default;
5. Validate All Input with Zod
const schema = z.object({ email: z.string().email() });
const validated = schema.parse(req.body);
6. Use Repository Pattern for Data Access
// Service → Repository → Database
const users = await userRepository.findActive();
7. Comprehensive Testing Required
describe('UserService', () => {
it('should create user', async () => {
expect(user).toBeDefined();
});
});
Common Imports
// Express
import express, { Request, Response, NextFunction, Router } from 'express';
// Validation
import { z } from 'zod';
// Database
import { PrismaClient } from '@prisma/client';
import type { Prisma } from '@prisma/client';
// Sentry
import * as Sentry from '@sentry/node';
// Config
import { config } from './config/unifiedConfig';
// Middleware
import { SSOMiddlewareClient } from './middleware/SSOMiddleware';
import { asyncErrorWrapper } from './middleware/errorBoundary';
Quick Reference
HTTP Status Codes
| Code | Use Case |
|---|---|
| 200 | Success |
| 201 | Created |
| 400 | Bad Request |
| 401 | Unauthorized |
| 403 | Forbidden |
| 404 | Not Found |
| 500 | Server Error |
Service Templates
Blog API (✅ Mature) - Use as template for REST APIs Auth Service (✅ Mature) - Use as template for authentication patterns
Anti-Patterns to Avoid
❌ Business logic in routes ❌ Direct process.env usage ❌ Missing error handling ❌ No input validation ❌ Direct Prisma everywhere ❌ console.log instead of Sentry
Navigation Guide
| Need to... | Read this |
|---|---|
| Understand architecture | architecture-overview.md |
| Create routes/controllers | routing-and-controllers.md |
| Organize business logic | services-and-repositories.md |
| Validate input | validation-patterns.md |
| Add error tracking | sentry-and-monitoring.md |
| Create middleware | middleware-guide.md |
| Database access | database-patterns.md |
| Manage config | configuration.md |
| Handle async/errors | async-and-errors.md |
| Write tests | testing-guide.md |
| See examples | complete-examples.md |
Resource Files
architecture-overview.md
Layered architecture, request lifecycle, separation of concerns
routing-and-controllers.md
Route definitions, BaseController, error handling, examples
services-and-repositories.md
Service patterns, DI, repository pattern, caching
validation-patterns.md
Zod schemas, validation, DTO pattern
sentry-and-monitoring.md
Sentry init, error capture, performance monitoring
middleware-guide.md
Auth, audit, error boundaries, AsyncLocalStorage
database-patterns.md
PrismaService, repositories, transactions, optimization
configuration.md
UnifiedConfig, environment configs, secrets
async-and-errors.md
Async patterns, custom errors, asyncErrorWrapper
testing-guide.md
Unit/integration tests, mocking, coverage
complete-examples.md
Full examples, refactoring guide
Related Skills
- database-verification - Verify column names and schema consistency
- error-tracking - Sentry integration patterns
- skill-developer - Meta-skill for creating and managing skills
Skill Status: COMPLETE ✅ Line Count: < 500 ✅ Progressive Disclosure: 11 resource files ✅
Signals
- GitHub stars
- 32k
- Forks
- 4k
- Last commit
- Sep 2026
Questions
- When should the agent use this skill?
- When creating routes, controllers, services, repositories, or middleware, or when working with Express APIs, Prisma database access, Sentry error tracking, Zod validation, unifiedConfig, dependency injection, or async code.
- What codebases does it apply to?
- Node.js, Express, and TypeScript microservices.
- Does it change how my agent writes code?
- Yes. It makes the agent follow the guidelines' house rules for backend service code instead of choosing its own patterns.
Advanced
- Item type
- skill
- Key
backend-dev-guidelines-davila7- Source
- github.com/davila7/claude-code-templates