Same .NET code. Fewer tokens.

same .NET code · fewer tokens · no magic

Smoower.Minified is a set of small C# libraries that strip the boilerplate out of .NET APIs (controllers, EF Core queries, DI, logging) and replace it with short, stable forms an AI can emit in a fraction of the tokens. It's plain C#: same IL, no transpiler.

.NET 8 · 9 · 10 no transpiler · same IL source-available ships as a Claude Code skill

What it is

When an AI writes ASP.NET Core or EF Core code, most of what it types isn't your logic. It's framework ceremony: Task<IActionResult>, FirstOrDefaultAsync, AddScoped, attribute noise, the same controller scaffolding over and over. Every token of it is regenerated on each write, rewrite, and refactor.

Smoower.Minified swaps that ceremony for short, stable helpers (Tr, .one(), scoped<>, [HG], .ok1()) that compile to the exact same IL. The alias layer is plain C#: extension methods, attributes, and type aliases. An opt-in [Crud<>] source generator and the deeper levels go further when you want them. The behavior never changes. Only the token count does.

Why it helps

Faster
generation time tracks output length. Less to type, less to wait for.
Cheaper
output tokens are billed, and there are simply fewer of them
Lighter
shorter code leaves more room in the context window over a session

Expect 10-25% fewer output tokens across a whole project, and up to 50% on the boilerplate-heavy controller files an assistant rewrites most often. The full arithmetic is on Does it pay off?

And it compounds with every agent step. A coding assistant doesn't write a file once. It works in a loop: read the code, edit it, re-read it, refactor, run again. Each step re-emits and re-reads the same source, so a token saved isn't saved once. It's saved again on every turn that touches that code, and again as that smaller output becomes the input the next step reads. The leaner each step, the leaner the one built on top of it, so the gap against the verbose form widens the longer an agent runs.

What it saves on real code

Not a toy benchmark. These are untouched production files from a live .NET app, rewritten in the compact style and measured with Claude’s own tokenizer (claude-opus-4-8). On files that are mostly framework ceremony, the compact form cuts up to 50% of the tokens. Across a whole application, where most of the code is business logic that shouldn't compress, the saving settles around 25%.

FileOriginalSmoowerSaved
AdminAuditController.cs7,3265,304−2,022 · 28%
EntryExpiryController.cs3,6402,510−1,130 · 31%
ExpiryDashboardController.cs4,0232,272−1,751 · 44%
RasepiDbContext.cs (~2k-line EF schema)31,71323,993−7,720 · 24%
Total46,70234,079−12,623 · 27%

At Opus 4.8’s rate of $25 per million output tokens, emitting this sample the verbose way costs $1.17. The compact way costs $0.85. That's $0.32 saved every time these files are written, and AI-generated code gets written, rewritten, and refactored over and over across a project's life.

The one rule: never compact the contract. Route templates, HTTP verbs, status codes, and DTO/JSON names stay exactly as your API requires. Smoower changes how code is written, never what it does at runtime.

Where to next