ruby-architect
Architect-level Ruby guidance — application layering beyond fat models and controllers, service and query object boundaries, ActiveRecord usage limits, background job and connection-pool discipline, and the concurrency reality of a Ruby service. Use when designing or reviewing a Ruby (Rails or Rack) backend.
- License 许可
- MIT
- 最近更新
- 2026-11-13
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- ruby-architect
- license
- MIT
- metadata
- { "crusor-path": "website-architect/backend-development-architect/ruby-architect", "crusor-domain": "website-architect", "crusor-origin": "original", "crusor-sources": "guides.rubyonrails.org; api.rubyonrails.org; shopify.github.io/ruby-style-guide", "crusor-triggers": "ruby, rails, activerecord, sidekiq, rack, ruby on rails, service objects", "crusor-verified-on": "2026-10-30" }
- description
- Architect-level Ruby guidance — application layering beyond fat models and controllers, service and query object boundaries, ActiveRecord usage limits, background job and connection-pool discipline, and the concurrency reality of a Ruby service. Use when designing or reviewing a Ruby (Rails or Rack) backend.
- compatibility
- Ruby 3.2+; Rails 7/8 or Sinatra/Rack
03 SKILL.md 正文
Ruby Architect
Ruby's flexibility is the risk: without explicit boundaries, business logic dissolves into models, controllers and callbacks.
When to use
- Rails product development, internal tools, API backends.
- Existing Ruby codebase whose models have 900 lines.
When NOT to use
- CPU-bound services or high-concurrency fan-out (Ruby's thread model and GVL make other runtimes cheaper).
- Edge/serverless workloads with cold-start constraints.
Decision 1 — Layering
app/controllers/ params, auth, response codes
app/services/ use-cases, transactions
app/queries/ read paths (query objects)
app/models/ persistence + small domain behaviour
- A service object takes plain arguments, returns a result, raises typed errors.
- Controllers never contain business rules; models never call external APIs.
- Callbacks are for persistence invariants only — anything else is a hidden action that will surprise someone.
concerns/only for genuinely shared behaviour; it is the most common hiding place for coupling.
Decision 2 — ActiveRecord discipline
- Scopes for reusable filters; never build queries in views or controllers.
- Eager-load (
includes/preload) deliberately; N+1 is the default failure. - Keep transactions in services, short; no HTTP calls inside them.
find_bywith explicit nil handling; avoidrescue nilandtrychains that hide state.- Migrations reversible or explicitly forward-only; no data logic in schema migrations.
Decision 3 — Concurrency and runtime
- Puma workers × threads sized to the DB pool — the pool is the real limit.
- Assume the GVL: CPU-heavy work belongs in a job or a separate process.
- Background jobs (Sidekiq) are idempotent and safe to retry; no job depends on a transient in-memory state.
- Timeouts: DB statement timeout, HTTP client timeouts, job timeout.
- Memory bloat: monitor RSS; long-running processes need restart policies.
Decision 4 — Testing and quality
- Request specs for boundaries, model/unit specs for rules; avoid testing Rails itself.
- Factories produce the minimum valid object; no shared mutable global state.
- RuboCop for style,
brakemanfor security,bundler-auditfor dependencies, all in CI.
Failure modes at scale
- N+1 queries discovered in production on the busiest page.
- A callback that sends email inside a transaction.
- Job retries creating duplicate side effects.
- Threads > DB pool → connection timeouts under load.
- A 1,500-line model with 40 public methods used nowhere else.
Verification receipts
- Query count per endpoint (no N+1) in request specs.
- Transaction duration on the busiest write path.
- Puma threads vs pool size documented and load-tested.
- Job idempotency test: run twice, one effect.
- Brakeman + bundler-audit clean in CI.
04 同仓库更多 Skills
interaction-patterns-architect
Architect-level interaction design in code — choosing established patterns over invented ones, feedback and latency behaviour, destructive-action safeguards, navigation and overlay conventions, and the state model behind every interactive surface. Use when a screen has to behave, not just look.
design-system-ops-architect
Architect-level operation of a design system in code — token pipelines and versioning, component API and variant policy, the catalogue as the contract, adoption metrics, and the governance that stops a system from forking into three. Use when a component library must survive several teams.
design-handoff-architect
Architect-level design handoff — what must be specified before implementation starts, tokens over raw values, responsive and content-limit contracts, state matrices, and the review loop that catches drift between design and code. Use when a design is about to become code.
content-microcopy-architect
Architect-level UI content — where strings live, voice and terminology consistency, error and empty-state copy as part of the interface, translation-ready authoring, and the content model a product needs before it multiplies screens. Use when the words in the interface matter as much as the pixels.
ui-ux-implementation-architect
Architect-level direction for turning a design into working UI — the handoff contract, component ownership versus screen-level layout, interaction and state conventions, design QA gates, and the review habits that keep an implementation faithful without frozen pixels. Use before any specific UI implementation skill.
test-strategy-architect
Architect-level test strategy — allocating assertions across unit, integration, contract and end-to-end layers, designing for testability through dependency boundaries, fixture and data strategy, and the flake and coverage policies that keep a suite trustworthy. Use when defining how a codebase is tested.