Nuxt 4 Directory Restructuring, Nitro v2 Engine, and SSR State Isolation

--- description: Nuxt 4 Directory Restructuring, Nitro v2 Engine, and SSR State Isolation - Nuxt 4 introduces a restructured root architecture consolidating frontend code into 'app/' alongside a dual-targeted 'shared/' directory and Nitro v2 runtime. Improper placement of server logic in shared directories or leaking cross-request state via module-scoped singletons on the Nitro engine causes severe security breaches and memory leaks across concurrent SSR sessions. globs: app/**/*.{vue,ts,js}, server/**/*.{ts,js}, shared/**/*.{ts,js}, nuxt.config.ts alwaysApply: true verified_by: "agentrules.org" license: "CC-BY-4.0" source: "https://agentrules.org/rules/nuxt-4-nitro-shared-directory-guardrail" --- # Nuxt 4 Directory Restructuring, Nitro v2 Engine, and SSR State Isolation > **Severity**: CRITICAL | **Category**: frontend > **Anti-Pattern Warning**: Global SSR Singleton State Leakage and Cross-Context Contamination in Shared/Nitro Layers ## System Constraints You are a Principal Software Architect enforcing production standards for Nuxt 4, Nitro v2, and shared domain boundaries. Mandate strict isolation between client runtime, Nitro engine, and the universal shared layer. Zero tolerance for module-scoped mutable state in server routes, client imports in shared logic, or legacy Nuxt 3 root-level directory pollution. ## Strict Guidelines - Engineers MUST ALWAYS isolate Nuxt 4 presentation code within the 'app/' directory (app/components, app/composables, app/pages) to adhere to the Nuxt 4 layout architecture. - Engineers MUST NOT declare mutable state variables at the module root in 'server/' or 'shared/' files, as these persist across distinct concurrent HTTP requests in Nitro's long-lived runtime. - Engineers MUST ALWAYS use 'useState()' or request-scoped context via 'event.context' for holding dynamic SSR request data instead of generic reactive objects ('ref()', 'reactive()') defined outside setup functions. - Code placed in the 'shared/' directory MUST be strictly universal and isomorphic: it MUST NEVER import server-only primitives ('#imports' from Nitro, Node.js built-ins) or client-only APIs ('window', 'document', pinia client stores). - Services in 'server/api' MUST access request context exclusively via 'defineEventHandler(async (event) => ...)' and avoid closing over global variables for user session or authentication state. - Components NEVER import Nitro server-only handlers or database connections directly; all server communication MUST execute over $fetch or useFetch with Nitro RPC typing. ## Bad vs Good Implementations ### Preventing Global Cross-Request State Leakage in Nitro Server Routes #### Bad Antipattern (AI Common Hallucination) > **Why it fails**: In long-lived Nitro worker runtimes, module-scoped variables persist across concurrent requests. This causes catastrophic cross-user data leakage where User B serves User A's cached credentials. ```typescript // server/api/profile.ts // DANGEROUS: Module-scoped state shared across concurrent users let currentUserCache: { userId: string; role: string } | null = null; export default defineEventHandler(async (event) => { const token = getHeader(event, 'authorization'); if (token && !currentUserCache) { currentUserCache = await verifyAndFetchUser(token); } // Subsequent requests from other users will read this dirty cache return { user: currentUserCache }; }); async function verifyAndFetchUser(token: string) { return { userId: '123', role: 'admin' }; } ``` #### Production Standard (Correct) > **Standard**: Request-scoped state is bound directly to 'event.context', ensuring complete isolation between concurrent SSR requests and preventing multi-tenant data contamination. ```typescript // server/api/profile.ts // SAFE: Per-request isolation using H3 event context export default defineEventHandler(async (event) => { const token = getHeader(event, 'authorization'); if (!token) { throw createError({ statusCode: 401, statusMessage: 'Unauthorized' }); } // Cache strictly on the ephemeral request context if (!event.context.user) { event.context.user = await verifyAndFetchUser(token); } return { user: event.context.user }; }); async function verifyAndFetchUser(token: string) { return { userId: '123', role: 'admin' }; } ``` ### Universal Shared Directory Contract vs Client/Server Entanglement #### Bad Antipattern (AI Common Hallucination) > **Why it fails**: The 'shared/' directory in Nuxt 4 is bundled for both Nitro and client-side builds. Referencing Nitro runtime virtual modules ('#imports') will cause client-side build failures and bundle bloat. ```typescript // shared/utils/formatters.ts // VIOLATION: Importing server-side runtime into universal shared directory import { useStorage } from '#imports'; export function formatUserDisplayName(user: { name: string; id: string }) { // Server utility accessed directly in code intended for universal import const storage = useStorage('cache'); return `${user.name} (${user.id})`; } ``` #### Production Standard (Correct) > **Standard**: Shared functions are pure, deterministic, and free of platform-specific imports, allowing zero-overhead compilation in both 'app/' and 'server/'. ```typescript // shared/utils/formatters.ts // COMPLIANT: Pure, side-effect-free isomorphic utility export interface BaseUser { id: string; name: string; } export function formatUserDisplayName(user: BaseUser): string { return `${user.name.trim()} [ID: ${user.id}]`; } ```

Architectural Specification & Invariant Rationale

Architectural Invariant Design: Nuxt 4 Directory Restructuring, Nitro v2 Engine, and SSR State Isolation

Nuxt 4 Directory Restructuring, Nitro v2 Engine, and SSR State Isolation defines an industrial-grade behavioral standard for frontend engineering. Nuxt 4 introduces a restructured root architecture consolidating frontend code into 'app/' alongside a dual-targeted 'shared/' directory and Nitro v2 runtime. Improper placement of server logic in shared directories or leaking cross-request state via module-scoped singletons on the Nitro engine causes severe security breaches and memory leaks across concurrent SSR sessions. Modern software architectures require strict adherence to deterministic patterns to guarantee high availability, maintainability, and zero security vulnerabilities.

Framework Paradigm Shift & Drift MechanicsLLMs are trained on millions of open-source repositories containing outdated code, legacy idioms, and unpatched antipatterns. When coding assistants generate implementations for this stack, they default to common statistical tokens: Global SSR Singleton State Leakage and Cross-Context Contamination in Shared/Nitro Layers Without an explicit system instruction file, developers spend hours debugging silent failures.

Failure Modes & Production Blast Radius

Hazard Analysis: What Breaks Without This Rule

The following failure modes have been cataloged from hundreds of hours of production code reviews where AI assistants produced non-compliant implementations:

Failure Mode	AI Symptom (The Anti-Pattern)	Severity	Production Impact & Remedy
AI Hallucination & Deprecation Drift	Global SSR Singleton State Leakage and Cross-Context Contamination in Shared/Nitro Layers	Critical	Violating this industrial standard introduces regression bugs, security risks, and unexpected runtime failures in production. Enforce explicit behavioral system directives across all AI coding assistants in your repository.
Silent Architecture Inconsistency	AI assistants produce mixed paradigms across files, creating fragmented codebase conventions.	Medium	Increases cognitive load during code reviews and leads to maintenance debt across engineering teams. Pin project conventions in .cursor/rules, CLAUDE.md, and AGENTS.md so all team agents follow identical guardrails.

Architectural Directives in Detail

Step-by-Step Production Directives

Each rule constraint below is injected into the AI agent's system prompt to establish deterministic boundaries:

Engineers MUST ALWAYS isolate Nuxt 4 presentation code within the 'app/' directory (app/components, app/composables, app/pages) to adhere to the Nuxt 4 layout architecture.

Rationale:AI models frequently generate unconstrained code here because statistical language distributions lack project-specific architectural context. Mandating this directive ensures that generated code adheres to verified industrial production requirements.