frontend-large-feature-architecture

Verwenden Sie beim Erstellen, Ändern oder Refaktorieren großer Langfuse-Frontend-Funktionen, virtualisierter Listen, großer Tabellen, Controller-Komponenten, lokaler Funktionen…

npx skills add https://github.com/langfuse/langfuse --skill frontend-large-feature-architecture

Frontend Large Feature Architecture

Use this skill when building, changing, or refactoring a large frontend surface.

In this skill, "controller" means a component or hook that owns feature logic: data fetching, view state, table/list state, effects, actions, and expensive rendering. The problem is not the name; it is one place owning too many changing responsibilities.

Big Feature Rules

When a feature grows, split rendering from logic. Most components should be view-only. Data preparation should be pure. Complex user actions should live in external async functions or local-store actions. Effects are integration boundaries, not the normal way to derive state.

For the full rules, read references/big-feature-rules.md.

React Without useEffect

UI is a pure function of state. Do not use useEffect to derive, prepare, or sync data:

  • Derived values are computed in render (const title = data?.name ?? ""), never mirrored into state by an effect. Client state that refers to server data stores only the user's intent (an id) and derives the effective value by merging with query data in render.
  • When loaded data seeds editable state, split the component: an outer data-preparer/controller fetches and renders a loading state (prefer a skeleton; <Spinner /> as minimal fallback); the inner component receives the loaded value as an initialValue prop and seeds useState(initialValue). Do not render UI before its data is ready.
  • Define a form only where all initial values are already prepared; if data is still loading, the form lives deeper in the tree.
  • Complex actions live outside React as plain functions using the store or query client — not inside components, not behind effects.
  • useCallback/useMemo are premature optimization; fix re-render problems by splitting components, not memoizing.
  • useEffect is legitimate only for DOM/browser API integration: no (or minimal) dependencies, one concern, and a cleanup function.

For the golden example and full reasoning, read references/react-without-useeffect.md.

Required Model

  • The page/view owns lifecycle and creates feature-scoped dependencies.
  • Server/query state stays in tRPC/React Query; route state stays in the router/filter hooks.
  • High-frequency feature UI state belongs in a per-mount local vanilla Zustand store. Create it with lazy useState, not useMemo.
  • Global stores are only for truly cross-feature, cross-route product state.
  • React context may provide a stable store or action owner. Do not put frequently changing state directly in provider values.
  • Rendered rows/cells/items should be view-only or narrow containers. Put effects, subscriptions, data loading, and workflows outside expensive views.
  • Shared src/components/* exports should stay context-free or receive explicit props. Put view-scoped Zustand consumers in src/features/*.
  • Large feature folders should have a concise README.md owner map.
  • Migrate real features toward this model whenever you touch them: improve state boundaries, action workflows, data-preparation seams, and render boundaries. Hundreds of legacy cases remain — do not preserve a legacy shape just to keep a change small.

Local Store Default

Prefer a local vanilla Zustand store for large or high-frequency feature state. The store is created by the page/view and destroyed on unmount.

Use selectors that return primitives or stable references. If a component needs multiple values, use shallow selector helpers or split subscriptions so one changing field does not rerender unrelated UI.

Complex user workflows should live in actions/*.ts files or store actions. The component wires hooks and passes dependencies; the action owns the workflow.

For a complete effect audit or component-splitting recipes, use ../refactor-react-effects/SKILL.md.

When To Read References

Mehr Skills von langfuse

clickhouse-best-practices
langfuse
MUSS verwendet werden, wenn ClickHouse-Schemata, Abfragen oder Konfigurationen überprüft werden. Enthält 28 Regeln, die vor der Abgabe von Empfehlungen geprüft werden MÜSSEN. Immer lesen…
official
skill-creator
langfuse
Leitfaden zur Erstellung effektiver Skills. Dieser Skill sollte verwendet werden, wenn Benutzer einen neuen Skill erstellen (oder einen bestehenden Skill aktualisieren) möchten, der Claudes…
official
vercel-react-best-practices
langfuse
Richtlinien zur Leistungsoptimierung von React und Next.js von Vercel Engineering. Diese Fähigkeit sollte beim Schreiben, Überprüfen oder Refaktorieren von React/Next.js…
official
add-model-price
langfuse
Use when editing worker/src/constants/default-model-prices.json, packages/shared/src/server/llm/types.ts, pricing tiers, tokenizer IDs, or matchPattern regexes…
official
analyze-cloud-costs
langfuse
Analysieren Sie die Kostenstruktur der Langfuse-Cloud-Infrastruktur mithilfe von Metabase-Kosten-Marts. Verwenden Sie dies, wenn nach Cloud-Ausgaben, AWS- versus ClickHouse-Kostenaufteilungen, Kosten… gefragt wird.
official
backend-dev-guidelines
langfuse
Build or review Langfuse backend code. Use for tRPC routers, public REST APIs, BullMQ processors, services, middleware, Prisma or ClickHouse access,…
official
clickhouse-best-practices
langfuse
MUSS verwendet werden, wenn ClickHouse-Schemas, Abfragen oder Konfigurationen überprüft werden. Enthält 28 Regeln, die vor der Abgabe von Empfehlungen geprüft werden MÜSSEN. Immer lesen…
official
code-review
langfuse
Review Langfuse code changes for correctness, regressions, and best practices.
official