clean-architecture · codewithmukesh — Claude Code skill for .NET layers
A skill that lays out .NET code across four projects, keeps dependencies pointing inward, and shows where use case handlers belong.
npx skills add codewithmukesh/dotnet-claude-kit --skill clean-architecture -a claude-codeWhat clean-architecture does
This skill covers Clean Architecture for .NET: Domain, Application, Infrastructure and Api as separate projects. Dependencies point inward, and the compiler enforces that through project references. Domain holds business rules as entity methods and knows nothing about databases or HTTP — just C# and .NET primitives.
You get code templates, not just prose. A minimal IAppDbContext abstraction instead of a repository layer, command and query handlers built on Mediator, validators with FluentValidation, entities with behaviour and a static Create factory. Infrastructure is treated as a plugin: EF Core, email senders and external APIs implement interfaces declared in Application or Domain, so swapping one out leaves business logic alone. The Api layer stays thin — endpoints only translate HTTP into use case calls.
The templates assume a specific stack: source-generated Mediator, FluentValidation, EF Core. Bring a different mediator library, or keep repositories on top of DbSet on purpose, and you will be rewriting the samples to fit.
What it helps with
- split a .NET solution into Domain, Application, Infrastructure and Api
- move business rules into entities instead of fat service classes
- give each use case its own command or query handler
- wire EF Core and external services through interfaces without touching logic
- keep API endpoints free of business code
Example prompts for the agent
- Scaffold a Clean Architecture project layout for orders
- Add a CreateOrder command with its handler and validator
- Rewrite this service as a use case handler using IAppDbContext
How to install clean-architecture
1Skills CLI (Vercel)
npx skills add codewithmukesh/dotnet-claude-kit --skill clean-architecture -a claude-code2Manually: put the files in place
~/.claude/skills/clean-architecture/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/clean-architecture/
1Skills CLI (Vercel)
npx skills add codewithmukesh/dotnet-claude-kit --skill clean-architecture -a codex2Manually: put the files in place
.agents/skills/clean-architecture/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add codewithmukesh/dotnet-claude-kit --skill clean-architecture -a cursor2Manually: put the files in place
.cursor/skills/clean-architecture/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install codewithmukesh/dotnet-claude-kit clean-architecture2Manually: put the files in place
.github/skills/clean-architecture/SKILL.md1Skills CLI (Vercel)
npx skills add codewithmukesh/dotnet-claude-kit --skill clean-architecture -a gemini-cli2Platform CLI
gemini skills install https://github.com/codewithmukesh/dotnet-claude-kit --path skills/clean-architectureManually: put the files in place
.devin/skills/clean-architecture/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add codewithmukesh/dotnet-claude-kit --skill clean-architecture -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about clean-architecture
How do I install clean-architecture in Claude?
Run npx skills add codewithmukesh/dotnet-claude-kit --skill clean-architecture -a claude-code. The skill file lands at ~/.claude/skills/clean-architecture/SKILL.md.
What does clean-architecture do?
It describes the four-project layout for .NET and supplies templates: the IAppDbContext abstraction, command and query handlers, validators, and entities with behaviour. The skill loads when you design layers or discuss dependency inversion.
Which stack does the skill assume?
The samples use source-generated Mediator, FluentValidation and EF Core. IAppDbContext is preferred over a repository, on the grounds that DbSet already plays that role.
Source text
SKILL.md · MIT
---
name: clean-architecture
description: >
Clean Architecture for .NET applications. Covers the 4-project layout (Domain,
Application, Infrastructure, Api), dependency inversion, use case handlers,
domain entities with behavior, and infrastructure as a plugin.
Load this skill when building a project with Clean Architecture, discussing
layered architecture, dependency inversion, use cases, or when the
architecture-advisor recommends Clean Architecture.
---
# Clean Architecture
## Core Principles
1. **Dependency inversion is the foundation** — All dependencies point inward. Domain has zero project references. Application references only Domain. Infrastructure references Application and Domain. Api references all but depends on abstractions. The compiler enforces this via project references.
2. **Domain owns the rules** — Business logic lives in the Domain layer as entity methods, domain services, or specifications. The Domain layer has no knowledge of databases, HTTP, or any framework — only pure C# and .NET primitives.
3. **Use cases are the unit of work** — Each use case (command or query) is a single class in the Application layer. It orchestrates domain objects, persists through abstractions, and returns a result. No "service" classes with 20 methods.
4. **Infrastructure is a plugin** — EF Core, external APIs, email senders, file storage — all live in Infrastructure and implement interfaces defined in Application or Domain. Swap implementations without touching business logic.
5. **The API layer is thin** — Endpoints map HTTP to use cases and use cases to HTTP responses. No business logic in endpoints.
## Patterns
### Project Layout
```
src/
MyApp.Domain/
Entities/
Order.cs # Entity with behavior
OrderItem.cs
Enums/
OrderStatus.cs
Exceptions/
DomainException.cs # Base domain exception
Interfaces/
IOrderRepository.cs # Only if query needs go beyond DbSet
Common/
Entity.cs # Base entity with Id
Result.cs # Result pattern type
MyApp.Application/
Common/
Behaviors/
ValidationBehavior.cs # Mediator pipeline behavior
Interfaces/
IAppDbContext.cs # DbContext abstraction (preferred over repository)
Orders/
Commands/
CreateOrder/
CreateOrderCommand.cs
CreateOrderHandler.cs
CreateOrderValidator.cs
Queries/
GetOrder/
GetOrderQuery.cs
GetOrderHandler.cs
OrderDto.cs
MyApp.Infrastructure/
Persistence/
AppDbContext.cs # Implements IAppDbContext
Configurations/
OrderConfiguration.cs
Migrations/
Services/
EmailSender.cs # Implements IEmailSender from Application
DependencyInjection.cs # AddInfrastructure extension
MyApp.Api/
Endpoints/
OrderEndpoints.cs # Thin, maps HTTP ↔ use cases
Program.cs
```
### DbContext Abstraction (Preferred Over Repository)
Define a minimal interface in Application; implement in Infrastructure:
```csharp
// Application/Common/Interfaces/IAppDbContext.cs
public interface IAppDbContext
{
DbSet<Order> Orders { get; }
DbSet<Product> Products { get; }
Task<int> SaveChangesAsync(CancellationToken ct = default);
}
// Infrastructure/Persistence/AppDbContext.cs
public class AppDbContext(DbContextOptions<AppDbContext> options)
: DbContext(options), IAppDbContext
{
public DbSet<Order> Orders => Set<Order>();
public DbSet<Product> Products => Set<Product>();
protected override void OnModelCreating(ModelBuilder modelBuilder)
{
modelBuilder.ApplyConfigurationsFromAssembly(typeof(AppDbContext).Assembly);
}
}
```
Why IAppDbContext over IRepository? EF Core's DbSet already IS a repository. Adding another abstraction on top adds indirection without value in most cases.
### Use Case Handler (Command)
```csharp
// Application/Orders/Commands/CreateOrder/CreateOrderCommand.cs
public record CreateOrderCommand(
string CustomerId,
List<OrderItemDto> Items) : IRequest<Result<Guid>>;
public record OrderItemDto(string ProductId, int Quantity, decimal UnitPrice);
// Application/Orders/Commands/CreateOrder/CreateOrderHandler.cs — uses Mediator (source-generated, MIT)
internal sealed class CreateOrderHandler(
IAppDbContext db,
TimeProvider clock) : IRequestHandler<CreateOrderCommand, Result<Guid>>
{
public async ValueTask<Result<Guid>> Handle(CreateOrderCommand request, CancellationToken ct)
{
var order = Order.Create(
request.CustomerId,
request.Items.Select(i => new OrderItem(i.ProductId, i.Quantity, i.UnitPrice)),
clock.GetUtcNow());
db.Orders.Add(order);
await db.SaveChangesAsync(ct);
return Result.Success(order.Id);
}
}
// Application/Orders/Commands/CreateOrder/CreateOrderValidator.cs
public class CreateOrderValidator : AbstractValidator<CreateOrderCommand>
{
public CreateOrderValidator()
{
RuleFor(x => x.CustomerId).NotEmpty();
RuleFor(x => x.Items).NotEmpty();
RuleForEach(x => x.Items).ChildRules(item =>
{
item.RuleFor(x => x.ProductId).NotEmpty();
item.RuleFor(x => x.Quantity).GreaterThan(0);
item.RuleFor(x => x.UnitPrice).GreaterThan(0);
});
}
}
```
### Use Case Handler (Query)
```csharp
// Application/Orders/Queries/GetOrder/GetOrderQuery.cs
public record GetOrderQuery(Guid OrderId) : IRequest<Result<OrderDto>>;
public record OrderDto(Guid Id, string CustomerId, decimal Total, string Status, DateTimeOffset CreatedAt);
// Application/Orders/Queries/GetOrder/GetOrderHandler.cs
internal sealed class GetOrderHandler(IAppDbContext db) : IRequestHandler<GetOrderQuery, Result<OrderDto>>
{
public async ValueTask<Result<OrderDto>> Handle(GetOrderQuery request, CancellationToken ct)
{
var order = await db.Orders
.Where(o => o.Id == request.OrderId)
.Select(o => new OrderDto(o.Id, o.CustomerId, o.Total, o.Status.ToString(), o.CreatedAt))
.FirstOrDefaultAsync(ct);
return order is not null
? Result.Success(order)
: Result.Failure<OrderDto>("Order not found");
}
}
```
### Domain Entity with Behavior
```csharp
// Domain/Entities/Order.cs
public class Order : Entity
{
private readonly List<OrderItem> _items = [];
private Order() { } // EF Core
public string CustomerId { get; private set; } = null!;
public OrderStatus Status { get; private set; }
public decimal Total { get; private set; }
public DateTimeOffset CreatedAt { get; private set; }
public IReadOnlyList<OrderItem> Items => _items.AsReadOnly();
public static Order Create(string customerId, IEnumerable<OrderItem> items, DateTimeOffset now)
{
var order = new Order
{
Id = Guid.CreateVersion7(),
CustomerId = customerId,
Status = OrderStatus.Pending,
CreatedAt = now
};
foreach (var item in items)
order.AddItem(item);
return order;
}
public void AddItem(OrderItem item)
{
_items.Add(item);
Total = _items.Sum(i => i.Quantity * i.UnitPrice);
}
public Result Cancel()
{
if (Status is not OrderStatus.Pending)
return Result.Failure("Only pending orders can be cancelled");
Status = OrderStatus.Cancelled;
return Result.Success();
}
}
```
### Thin Endpoint Wiring (IEndpointGroup Auto-Discovery)
Every endpoint group implements `IEndpointGroup` and is auto-discovered via `app.MapEndpoints()`. Program.cs never changes when adding new endpoints. See the **minimal-api** skill for the full `IEndpointGroup` interface and `EndpointExtensions` setup.
```csharp
// Api/Endpoints/OrderEndpoints.cs
public sealed class OrderEndpoints : IEndpointGroup
{
public void Map(IEndpointRouteBuilder app)
{
var group = app.MapGroup("/api/orders").WithTags("Orders");
group.MapPost("/", CreateOrder)
.WithName("CreateOrder");
group.MapGet("/{id:guid}", GetOrder)
.WithName("GetOrder");
group.MapGet("/", ListOrders)
.WithName("ListOrders");
}
private static async Task<IResult> CreateOrder(
CreateOrderCommand command, ISender sender, CancellationToken ct)
{
var result = await sender.Send(command, ct);
return result.IsSuccess
? TypedResults.Created($"/api/orders/{result.Value}", result.Value)
: result.ToProblemDetails();
}
private static async Task<IResult> GetOrder(
Guid id, ISender sender, CancellationToken ct)
{
var result = await sender.Send(new GetOrderQuery(id), ct);
return result.IsSuccess
? TypedResults.Ok(result.Value)
: TypedResults.NotFound();
}
private static async Task<IResult> ListOrders(
[AsParameters] ListOrdersQuery query, ISender sender, CancellationToken ct)
{
var result = await sender.Send(query, ct);
return TypedResults.Ok(result);
}
}
```
### Infrastructure DI Registration
```csharp
// Infrastructure/DependencyInjection.cs
public static class DependencyInjection
{
public static IServiceCollection AddInfrastructure(
this IServiceCollection services,
IConfiguration config)
{
services.AddDbContext<AppDbContext>(options =>
options.UseNpgsql(config.GetConnectionString("DefaultConnection")));
services.AddScoped<IAppDbContext>(sp => sp.GetRequiredService<AppDbContext>());
return services;
}
}
```
## Anti-patterns
### Anemic Domain Model
```csharp
// BAD — entity is just a data bag, all logic in handler
public class Order
{
public Guid Id { get; set; }
public string CustomerId { get; set; } = null!;
public decimal Total { get; set; }
public List<OrderItem> Items { get; set; } = [];
}
// Handler sets everything directly
order.Total = order.Items.Sum(i => i.Quantity * i.UnitPrice);
order.Status = OrderStatus.Pending;
// GOOD — entity encapsulates its own rules (see Domain Entity pattern above)
var order = Order.Create(customerId, items, clock.GetUtcNow());
```
### DbContext in Domain Layer
```csharp
// BAD — Domain references EF Core
// Domain/Services/OrderService.cs
public class OrderService(AppDbContext db) { } // Domain depends on Infrastructure!
// GOOD — Domain defines interfaces, Infrastructure implements
// Domain/Interfaces/IOrderRepository.cs (only if you need query abstraction beyond DbSet)
// Application/Common/Interfaces/IAppDbContext.cs (preferred)
```
### Fat Endpoints
```csharp
// BAD — business logic in the endpoint
app.MapPost("/orders", async (CreateOrderRequest req, AppDbContext db) =>
{
var order = new Order { CustomerId = req.CustomerId };
foreach (var item in req.Items)
{
order.Items.Add(new OrderItem { ProductId = item.ProductId, Quantity = item.Quantity });
}
order.Total = order.Items.Sum(i => i.Quantity * i.UnitPrice);
db.Orders.Add(order);
await db.SaveChangesAsync();
return TypedResults.Created($"/orders/{order.Id}", order);
});
// GOOD — endpoint delegates to a use case
app.MapPost("/orders", async (CreateOrderCommand command, ISender sender, CancellationToken ct) =>
{
var result = await sender.Send(command, ct);
return result.IsSuccess
? TypedResults.Created($"/orders/{result.Value}", result.Value)
: result.ToProblemDetails();
});
```
### Repository for Every Entity
```csharp
// BAD — repository per entity duplicates DbSet functionality
public interface IOrderRepository { Task<Order?> GetByIdAsync(Guid id); }
public interface IProductRepository { Task<Product?> GetByIdAsync(Guid id); }
public interface ICustomerRepository { Task<Customer?> GetByIdAsync(Guid id); }
// GOOD — use IAppDbContext with DbSet<T> directly
// Only create a repository interface when you have complex query logic
// that you want to test in isolation or reuse across multiple use cases
```
## Decision Guide
| Scenario | Recommendation |
|----------|---------------|
| When to use CA over VSA | Medium+ domain complexity, long-lived system, team familiar with layers |
| When to add a Domain layer | Business rules involve invariants across entity groups |
| IAppDbContext vs repositories | Prefer IAppDbContext; add repository only for complex reusable queries |
| Mediator vs raw handlers in CA | Mediator for pipeline behaviors (validation, logging); raw handlers for simplicity |
| When to add Domain events | When side effects (notifications, audit) should be decoupled from the main flow |
| Evolving from VSA to CA | When handlers start needing shared domain logic that does not belong in Common/ |
© codewithmukesh · MIT · codewithmukesh/dotnet-claude-kit
Similar skills
Everything for Backend and APIs →clean-architecture by other authors comes first.