NetSkelly

A skeleton .NET API project using Clean Architecture. It exposes a filter-set to the caller—any field on any object can be filtered on—while keeping the codebase focused.

---

NetSkelly is a .NET API skeleton project designed with Clean Architecture principles to provide a robust starting point for web services. It addresses the common need for flexible data querying by implementing a dynamic, URL-based filtering mechanism that allows clients to filter any field on returned objects using various operators.

The project also features a comprehensive authentication and authorization system, including multi-factor authentication (MFA) and JWT-based security.

Its modular design separates concerns into API, Application, Domain, and Infrastructure layers, promoting maintainability and scalability.

Notable Elements

• Custom Dynamic Query Filtering: Implemented a robust URL-based filtering system using a custom model binder and parser, enabling dynamic filtering on any object property with various operators, significantly reducing API endpoint boilerplate for data retrieval.
• Adherence to Clean Architecture: The project's explicit separation of concerns into API, Application, Domain, and Infrastructure projects demonstrates a commitment to modularity, testability, and long-term maintainability.
• Centralized API Exception Handling: Utilized an ASP.NET Core ExceptionFilter to convert application-layer exceptions into standardized HTTP responses, streamlining error management and ensuring a consistent API error contract.

generated by RepoLens

---

API Usage

imagine an endpoint https://net.skelly/record and a some "Record" objects as follows:

[
  {
    "id": 1,
    "name": "lorem recordum",
    "is_cool": true,
    "points": 12,
    "created_at": 2026-04-11 15:00,
  },
  {
    "id": 2,
    "name": "dorum et nihilus",
    "is_cool": false,
    "points": 15,
    "created_at": 2026-04-11 16:00,
  },
  {
    "id": 2,
    "name": "dali to pahealus",
    "is_cool": true,
    "points": 33,
    "created_at": 2026-04-11 14:00,
  },
]

The API user can then fetch a specific record with https://net.skelly/record/{id} as expected, but they can also list records using a set of filters which correspond to the properties on the objects returned.

This is most effectively done through the creation of a filters object on the calling code, so that the url will contain ?filters[foo]=bar. The obvious is an equals operation so it defaults to that. so for example if filtering for all things cool then ?filters[is_cool]=true would return the 2 items listed as cool.

this is done on the root "uri" like https://net.skell/record?filters[is_cool]=true It can also use the following operators:

• prop.eq -> equals
• prop.ne -> not equals
• prop.lt -> less than
• prop.lte -> less than or equal to
• prop.gt -> greater than
• prop.gte -> greater than or equal to
• prop.contains -> string comparison, returns all that contain the query

multiple filters on the same property can also be passed which will be converted into a logical AND for lt, lte, gt, and gte and a logical OR for eq, ne, and contains.

see the EFRepository class for more details on listing with filters.