Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Provides comprehensive code review capability for NestJS applications, analyzing controllers, services, modules, guards, interceptors, pipes, dependency injection, and database integration patterns. Use when reviewing NestJS code changes, before merging pull requests, after implementing new features, or for architecture validation. Triggers on "review NestJS code", "NestJS code review", "check my NestJS controller/service".
.claude/skills/giuseppe-trisciuoglio-nestjs-code-review/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✓→✗ | ▼ Worse | -46% | 0% |
| case-19 | ✓→✗ | ▼ Worse | 69% | 0% |
| case-07 | ✓→✓ | = Same ✓ | 114% | 0% |
| case-02 | ✓→✓ | = Same ✓ | 78% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 28% | 0% |
Provides structured code review for NestJS applications. Findings categorized by severity (Critical, Warning, Suggestion) with actionable recommendations. Delegates to nestjs-code-review-expert agent for deep analysis.
glob and grep to discover controllers, services, modules, guards, interceptors, and pipes in the target area.typescript// ❌ Bad: Fat controller with business logic and missing validation @Controller('users') export class UserController { constructor(private readonly userRepo: Repository<User>) {} @Post() async create(@Body() body: any) { const user = this.userRepo.create(body); return this.userRepo.save(user); } } // ✅ Good: Thin controller with proper DTOs, validation, and service delegation @Controller('users') @ApiTags('Users') export class UserController { constructor(private readonly userService: UserService) {} @Post() @HttpCode(HttpStatus.CREATED) @ApiOperation({ summary: 'Create a new user' }) @ApiResponse({ status: 201, type: UserResponseDto }) async create( @Body(ValidationPipe) createUserDto: CreateUserDto, ): Promise<UserResponseDto> { return this.userService.create(createUserDto); } }
typescript// ❌ Bad: Direct instantiation bypasses DI @Injectable() export class OrderService { private readonly logger = new Logger(); private readonly emailService = new EmailService(); async createOrder(dto: CreateOrderDto) { this.emailService.send(dto.email, 'Order created'); } } // ✅ Good: Proper constructor injection @Injectable() export class OrderService { private readonly logger = new Logger(OrderService.name); constructor( private readonly orderRepository: OrderRepository, private readonly emailService: EmailService, ) {} async createOrder(dto: CreateOrderDto): Promise<Order> { const order = await this.orderRepository.create(dto); await this.emailService.send(dto.email, 'Order created'); return order; } }
typescript// ❌ Bad: Generic error handling with information leakage @Get(':id') async findOne(@Param('id') id: string) { try { return await this.service.findOne(id); } catch (error) { throw new HttpException(error.message, 500); } } // ✅ Good: Domain-specific exceptions with proper HTTP mapping @Get(':id') async findOne(@Param('id', ParseUUIDPipe) id: string): Promise<UserResponseDto> { const user = await this.userService.findOne(id); if (!user) { throw new NotFoundException(`User with ID ${id} not found`); } return user; }
typescript// ❌ Bad: Authorization logic in controller @Get('admin/dashboard') async getDashboard(@Req() req: Request) { if (req.user.role !== 'admin') { throw new ForbiddenException(); } return this.dashboardService.getData(); } // ✅ Good: Guard-based authorization with decorator @Get('admin/dashboard') @UseGuards(JwtAuthGuard, RolesGuard) @Roles(Role.ADMIN) async getDashboard(): Promise<DashboardDto> { return this.dashboardService.getData(); }
typescript// ❌ Bad: Monolithic module with everything @Module({ imports: [TypeOrmModule.forFeature([User, Order, Product, Review])], controllers: [UserController, OrderController, ProductController], providers: [UserService, OrderService, ProductService, ReviewService], }) export class AppModule {} // ✅ Good: Feature-based module organization @Module({ imports: [UserModule, OrderModule, ProductModule], }) export class AppModule {} @Module({ imports: [TypeOrmModule.forFeature([User])], controllers: [UserController], providers: [UserService, UserRepository], exports: [UserService], }) export class UserModule {}
Structure all code review findings as follows:
Brief overview with an overall quality score (1-10) and key observations.
Issues that could cause security vulnerabilities, data corruption, or production failures.
Issues that violate best practices, reduce maintainability, or could lead to bugs.
Improvements for code readability, performance, or developer experience.
Well-implemented patterns and good practices to acknowledge and encourage.
Prioritized next steps with code examples for the most impactful improvements.
ParseUUIDPipe, ParseIntPipe, etc. for parameter validationHttpExceptionnew for injectable services@ApiTags, @ApiOperation, @ApiResponse) to all endpointsSee the references/ directory for detailed review checklists and pattern documentation:
references/patterns.md — NestJS best practice patterns with examplesreferences/anti-patterns.md — Common NestJS anti-patterns to flag during reviewreferences/checklist.md — Comprehensive review checklist organized by area| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→fail | 40,096 | 5,181 | -87% | 1 | 1 | 0% | 3,856 | 2,101 | -46% | 0 | 0 | — |
case-07 | pass→pass | 9,344 | 10,479 | +12% | 1 | 1 | 0% | 1,783 | 3,812 | +114% | 0 | 0 | — |
case-02 | pass→pass | 12,853 | 11,415 | -11% | 1 | 1 | 0% | 2,347 | 4,166 | +78% | 0 | 0 | — |
case-03 | pass→pass | 13,233 | 6,308 | -52% | 1 | 1 | 0% | 2,498 | 3,191 | +28% | 0 | 0 | — |
case-04 | pass→pass | 12,676 | 11,670 | -8% | 1 | 1 | 0% | 2,222 | 4,041 | +82% | 0 | 0 | — |
case-05 | pass→pass | 11,333 | 10,252 | -10% | 1 | 1 | 0% | 1,430 | 3,811 | +167% | 0 | 0 | — |
case-06 | pass→pass | 11,238 | 10,341 | -8% | 1 | 1 | 0% | 1,943 | 3,907 | +101% | 0 | 0 | — |
case-08 | pass→pass | 12,455 | 8,070 | -35% | 1 | 1 | 0% | 1,959 | 3,340 | +70% | 0 | 0 | — |
case-09 | pass→pass | 7,496 | 8,501 | +13% | 1 | 1 | 0% | 1,254 | 3,385 | +170% | 0 | 0 | — |
case-10 | pass→pass | 13,019 | 11,661 | -10% | 1 | 1 | 0% | 2,440 | 3,975 | +63% | 0 | 0 | — |
case-11 | pass→pass | 12,503 | 9,904 | -21% | 1 | 1 | 0% | 2,160 | 3,642 | +69% | 0 | 0 | — |
case-12 | pass→pass | 11,897 | 11,544 | -3% | 1 | 1 | 0% | 2,252 | 4,134 | +84% | 0 | 0 | — |
case-13 | fail→fail | 8,219 | 5,252 | -36% | 1 | 1 | 0% | 1,503 | 2,164 | +44% | 0 | 0 | — |
case-14 | pass→pass | 13,586 | 9,444 | -30% | 1 | 1 | 0% | 2,553 | 3,835 | +50% | 0 | 0 | — |
case-15 | pass→pass | 9,954 | 25,487 | +156% | 1 | 1 | 0% | 1,745 | 3,808 | +118% | 0 | 0 | — |
case-16 | pass→pass | 9,707 | 7,629 | -21% | 1 | 1 | 0% | 1,470 | 3,076 | +109% | 0 | 0 | — |
case-17 | fail→fail | 20,379 | 4,704 | -77% | 1 | 1 | 0% | 3,799 | 2,115 | -44% | 0 | 0 | — |
case-18 | pass→pass | 10,334 | 12,583 | +22% | 1 | 1 | 0% | 1,691 | 4,047 | +139% | 0 | 0 | — |
case-19 | pass→fail | 14,149 | 11,430 | -19% | 1 | 1 | 0% | 2,484 | 4,206 | +69% | 0 | 0 | — |
case-20 | pass→pass | 9,517 | 9,892 | +4% | 1 | 1 | 0% | 1,925 | 3,824 | +99% | 0 | 0 | — |
case-21 | pass→pass | 32,480 | 22,645 | -30% | 1 | 1 | 0% | 3,686 | 6,778 | +84% | 0 | 0 | — |
case-22 | pass→pass | 11,024 | 13,134 | +19% | 1 | 1 | 0% | 2,169 | 4,413 | +103% | 0 | 0 | — |
case-23 | pass→pass | 11,770 | 7,338 | -38% | 1 | 1 | 0% | 2,292 | 3,273 | +43% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 20 counted toward the lift figure. The other 3 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -100 percentage points is the difference between those two pass rates over the 20 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.