npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@structure-ai/cqrs

v0.4.1

Published

Schema-typed command/query buses with validation, authorization, idempotency and tracing middleware.

Readme

@structure-ai/cqrs

Logical CQRS: Schema-typed commands and queries, one handler per message, buses as Effect services. Every dispatch goes through the same pipeline: boundary validation (shape only — business rules stay in the domain) → authorization → idempotency claim (commands with a key) → traced, measured handler execution.

Usage

import { Command, CommandBus, CommandHandler, HandlerRegistry, layer } from "@structure-ai/cqrs";
import { Effect, Schema } from "effect";

const ApproveInvoice = Command.define("ApproveInvoice", {
  payload: Schema.Struct({ invoiceId: Schema.String, approver: Schema.String }),
  success: Schema.Struct({ invoiceId: Schema.String, version: Schema.Number }),
});

const handlers = HandlerRegistry.layer(
  CommandHandler.make(ApproveInvoice, (payload, dispatch) =>
    Effect.succeed({ invoiceId: payload.invoiceId, version: 2 }),
  ),
);

const program = Effect.gen(function* () {
  const bus = yield* CommandBus;
  const ack = yield* bus.dispatch(ApproveInvoice, { invoiceId: "inv-1", approver: "ada" }, {
    idempotencyKey: "req-42",
  });
}).pipe(Effect.provide(layer.pipe(Layer.provide(handlers)))); // convenience layer: both buses, allow-all authorizer, in-memory idempotency

Exports

| Export | What it is | | --- | --- | | Command.define(tag, { payload, success, failure? }) / Query.define | Intent-named message definitions; failure types business failures; handlers may also fail with infrastructure PersistenceError. | | CommandHandler.make / QueryHandler.make | Bind a definition to a handler (payload, dispatch) => Effect; dispatch carries messageId, correlationId, causationId, actor, idempotencyKey. | | HandlerRegistry.layer(...registrations) | Collects handlers (duplicate tag = defect at build time); captures handler service requirements from the context at build — satisfy them with Layer.provideMerge, not Layer.provide, or the services are spent building the registry and missing from the runtime context (missing-service defects at dispatch, invisible to the type checker). | | CommandBus / QueryBus (+ .layer, convenience layer) | dispatch(definition, input, { idempotencyKey?, actor?, timeout? }). | | Authorizer (+ allowAll) | Hook authorizing the requested action (tag/kind/actor/payload), not the endpoint. | | IdempotencyStore (+ inMemory), IdempotencyStoreService, IdempotencyContext, BeginOutcome | Port for command idempotency: begin(context) claims (tag, actor, key) and returns Completed(result) / Claimed / InFlight / Mismatch; complete(context, result) records the wire-encoded success; release(context) frees a failed claim. inMemory is process-local; @structure-ai/eventsourcing-pg ships a durable one with TTL. | | HandlerNotFound, Unauthorized, DispatchTimeout, IdempotencyMismatch, IdempotencyInFlight | Tagged errors with classification; validation failures reuse ValidationFailed from @structure-ai/domain. |

Each dispatch runs in a span (cqrs.command.<Tag>), a metrics boundary (calls/errors/latency), and Correlation.within with causationId = messageId.

Idempotency

A command dispatched with idempotencyKey is identified by the key scoped to the acting principal (options.actor, falling back to the ambient correlation actor; anonymous is its own scope) and the command tag, and bound to a sha-256 of the validated payload in wire form (object keys sorted, so field order is irrelevant). Queries ignore the key.

| Situation | Outcome | | --- | --- | | First dispatch of the key | Store returns Claimed; the handler runs; its success is encoded with the definition's success schema and stored with complete. | | Same actor, same payload, again | Completed: the stored success is decoded and returned; the handler does not run (nor is anything traced). | | Same actor, same key, different payload | IdempotencyMismatch (classification: "conflict", HTTP 409): the handler does not run. | | Same actor, same key, first dispatch still running | IdempotencyInFlight (classification: "transient", HTTP 409): retry once it finished to get its result. | | Another actor, same key | Independent claim: the handler runs again. Keys never leak results across principals. | | The claimed dispatch fails, times out or is interrupted | The claim is released; the next dispatch runs the handler again. |

Stored results are the encoded success, so a durable store only ever holds JSON — make sure every field of a command's success schema round-trips (Schema.Date, branded ids and the like do). The store suppresses replays and duplicate concurrent runs of one key; exactly-once processing of events is the eventsourcing inbox's job.

Persistence failures

DispatchError includes PersistenceError from @structure-ai/domain. Handlers and IdempotencyStore.begin/complete/release preserve that typed failure. A stored result that no longer decodes is also a persistence failure. It stays separate from the public business failure schema and maps to a safe HTTP 500.

A failed dispatch releases its claim without interruption. If release also fails, the bus retains both causes; defects and interruption remain distinct. A failed completion can leave the claim in flight until its TTL expires. Do not automatically retry an ambiguous write.