Skuld
View SourceAn effectful programming framework for Elixir.
Comp
(lazy computation,
evidence-passing,
scoped handlers)
│
┌─────────────────────┼───────────────────────────┐
│ │ │
//Foundational //Coroutines & //Boundaries
//Effects //Concurrency │
│ │ │
│ Coroutine ┌──────┼────┐
│ │ │ │ │
State, Reader, ┌────────┼─────────┐ │ │ Port
Writer, Throw, │ │ │ │ │ Port.EffectfulFacade
Bracket, Fresh, │ Serializable- │ │ │ Repo
Random, FxList, │ Coroutine │ │ │
Yield, │ │ │ │
EffectLogger, AsyncCoroutine FiberPool │ Adapter
Parallel, │ │ Adapter.EffectfulContract
AtomicState, ├─────────┐│
Transaction, │ ││
Command │ ││
┌────┴────┐ ││
Channel Task ││
│ ││
Brook ││
││
Query.Contract
QueryBlock
(Haxl-like: auto-batches fetches
via Coroutine fibers)The old problem
Between pure business logic and side-effecting infrastructure sits the orchestration layer — "fetch the user, check permissions, load their subscription, hit some APIs, compute a price, write an invoice." This code encodes your most important business rules, but it's tangled with databases, APIs, and randomness — making it hard to test, hard to refactor, and often — impossible to property-test.
Another way
Skuld lets you write orchestration code that describes side effects without performing them — then handlers decide what those descriptions mean. The exact same "effectful" code runs with side-effecting handlers in production and pure in-memory handlers in tests — fully deterministic, fully pure, and straightforwardly property-testable.
Because effects are first-class data, Skuld can do more — batch independent queries automatically, serialise partially complete computations for later resumption.
Quick example
defmodule Onboarding do
use Skuld.Syntax
alias Skuld.Repo
alias Skuld.Effects.{Fresh, Reader, Writer}
defcomp register(params) do
config <- Reader.ask()
id <- Fresh.fresh_uuid()
{:ok, user} <- Repo.insert(User.changeset(%{id: id, name: params.name, tier: config.default_tier}))
_ <- Writer.tell(:events, %UserRegistered{user_id: id})
{:ok, user}
end
endRun with production handlers:
# One-time setup: generate an Ecto adapter for the Repo contract
defmodule MyApp.Repo.Port do
use Skuld.Repo.Ecto, repo: MyApp.Repo
end
# Wire everything up:
Onboarding.register(%{name: "Alice"})
|> Reader.with_handler(%{default_tier: :free})
|> Fresh.with_uuid7_handler()
|> Port.with_handler(%{Skuld.Repo.Effectful => MyApp.Repo.Port})
|> Writer.with_handler([], tag: :events, output: fn r, raw ->
MyApp.EventBus.publish(Enum.reverse(raw))
r
end)
|> Throw.with_handler()
|> Comp.run!()Run with test handlers — same code, fully deterministic, no database:
Onboarding.register(%{name: "Alice"})
|> Reader.with_handler(%{default_tier: :free})
|> Fresh.with_test_handler()
|> Repo.InMemory.with_handler(Repo.InMemory.new())
|> Writer.with_handler([], tag: :events, output: fn r, raw -> {r, Enum.reverse(raw)} end)
|> Throw.with_handler()
|> Comp.run!()Repo.InMemory is a closed-world in-memory store with read-after-write
consistency. Records created during the test are immediately readable by
subsequent Repo.get / Repo.get_by calls — no mocks, no stubs.
Installation
def deps do
[
{:skuld, "~> 0.27"}
]
endWhere next?
| If you want to... | Read |
|---|---|
| Understand the problem effects solve | Why Effects? |
| See how effects and handlers work | How It Works |
| Write your first computation | Getting Started |
| State, Reader, Writer, Throw, Fresh, Random | Foundational Effects |
| Yield, Coroutines, FiberPool, Channels, Async | Coroutines & Concurrency |
| Port, Repo, Hexagonal Architecture | Boundaries |
| Eliminate N+1 queries | Query System |
| Handler-swapping for deterministic testing | Testing |
| Full effect and API reference | Reference |
License
MIT License — see LICENSE for details.