TenantAtlas/specs/404-public-content-messaging/plan.md

Implementation Plan: Public Website Positioning & Content Architecture

Branch: 404-public-content-messaging | Date: 2026-05-25 | Spec: /Users/ahmeddarrazi/Documents/projects/wt-website/specs/404-public-content-messaging/spec.md Input: Feature specification from /Users/ahmeddarrazi/Documents/projects/wt-website/specs/404-public-content-messaging/spec.md

Summary

Reposition the public Tenantial website from an Intune-only or backup-tool impression toward Policy Governance for Microsoft 365 and modern cloud environments, with Microsoft 365 as the first focus and provider-extensible language kept explicitly future-safe. The implementation will stay inside /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website and reuse the existing Astro route/content architecture: locale-keyed copy in src/data_files/site-copy.ts, thin route wrappers in src/pages, shared page components in src/components/pages, metadata through MainLayout and Meta, and the current Playwright smoke suite for route, link, claim, and metadata validation.

Technical Context

Language/Version: TypeScript 6.0.3, Astro 6.3.3, Node.js >=20.0.0, pnpm 10.33.0 Primary Dependencies: Astro, @astrojs/starlight, @astrojs/sitemap, @astrojs/mdx, Tailwind CSS v4, @tailwindcss/vite, Preline 4, Lenis, GSAP, Sharp, Playwright Storage: N/A - static website content and generated build output only; no database or product persistence Testing: Astro build via corepack pnpm build:website, existing Playwright smoke tests under /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/tests/smoke, targeted static claim scans Validation Lanes: website build, public smoke, manual browser review, static claim scan, whitespace check, apps/platform scope check Target Platform: Static Astro public website deployed from /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website, with German default routes and /en/... mirrors Project Type: Web - standalone Astro public website inside a monorepo Performance Goals: No body-level horizontal overflow on validated desktop/mobile routes; primary navigation and CTAs stay readable and reachable; metadata and canonical routes stay intentional Constraints: Runtime/source changes are scoped to apps/website; preserve root package script names, website package name @tenantatlas/website, WEBSITE_PORT, and apps/* workspace conventions; no apps/platform changes; no fake trust/provider claims; no placeholder links; no auth/API/database/runtime coupling Scale/Scope: Core public pages /, /platform, /pricing, /trust, /contact, legal pages, exposed docs routes, locale mirrors, navigation/footer surfaces, route metadata, and smoke expectations

UI / Surface Guardrail Plan

• Guardrail scope: no operator-facing surface change; public website positioning workflow only
• Native vs custom classification summary: custom Astro public website; no Filament/Blade/admin surface
• Shared-family relevance: public navigation, CTA language, metadata, docs exposure, and smoke helper patterns
• State layers in scope: localized static page content, navigation/footer data, route metadata, docs content, smoke expectations
• Audience modes in scope: public visitor, MSP evaluator, internal IT evaluator, DACH trust reviewer
• Decision/diagnostic/raw hierarchy plan: public copy stays decision-first for visitors; diagnostics and proof boundaries are explained plainly rather than exposed as raw runtime detail
• Raw/support gating plan: N/A - no operator support/raw evidence surface
• One-primary-action / duplicate-truth control: each primary route keeps one clear next step, typically contact or deeper product explanation, while repeated or competing CTA language is normalized
• Handling modes by drift class or surface: public claim, placeholder-link, and navigation drift are review-mandatory inside this feature; apps/platform drift is a hard stop
• Repository-signal treatment: website-source and website-smoke changes are expected; any platform/runtime drift is exception-required and out of scope
• Special surface test profiles: N/A - public website only
• Required tests or manual smoke: public smoke, static claim scan, and manual desktop/mobile browser review
• Exception path and spread control: none
• Active feature PR close-out entry: Smoke Coverage

Shared Pattern & System Fit

• Cross-cutting feature marker: yes
• Systems touched: /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/src/data_files/site-copy.ts, /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/src/data_files/constants.ts, /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/src/components/pages, /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/src/components/sections/navbar&footer, /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/src/components/Meta.astro, /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/src/content/docs, and /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/tests/smoke
• Shared abstractions reused: locale-keyed siteCopy, thin route wrappers in src/pages, MainLayout.astro, Meta.astro, localizeHref() and locale helpers, shared Navbar/Footer components, Playwright smoke helper patterns for forbidden claims and placeholder links
• New abstraction introduced? why?: none
• Why the existing abstraction was sufficient or insufficient: The current website already centralizes copy, navigation, metadata, and smoke assertions. Spec 404 needs a better narrative and stricter claim posture, not a new framework.
• Bounded deviation / spread control: no new abstraction; bounded cleanup of stale helpers such as German-only navigation.ts usage is allowed if needed to keep copy and route logic aligned

OperationRun UX Impact

• Touches OperationRun start/completion/link UX?: no
• Central contract reused: N/A
• Delegated UX behaviors: N/A
• Surface-owned behavior kept local: N/A
• Queued DB-notification policy: N/A
• Terminal notification path: N/A
• Exception path: none

Provider Boundary & Portability Fit

• Shared provider/platform boundary touched?: yes
• Provider-owned seams: public Microsoft 365 wording, Intune as one example domain, any roadmap/provider-direction examples in public copy
• Platform-core seams: none; no runtime platform contracts, provider contracts, or shared persistence truth change
• Neutral platform terms / contracts preserved: policy governance, cloud policy governance, managed environment, provider connection, policy evidence, drift detection, findings, exceptions, accepted risks, decision summary, audit trail, controlled recovery, provider readiness
• Retained provider-specific semantics and why: Microsoft 365 remains the first public focus because that is current product truth; Intune is retained only as one Microsoft 365 policy domain and not the umbrella category
• Bounded extraction or follow-up path: follow-up-spec for a broader public provider/domain taxonomy if future route or copy work needs a richer current-versus-planned matrix

Constitution Check

GATE: Must pass before Phase 0 research. Re-check after Phase 1 design.

• Inventory-first: PASS - no inventory, snapshots, backups, or source-of-truth runtime behavior changes
• Read/write separation: PASS - no write/change behavior is introduced
• Graph contract path: PASS - no Microsoft Graph calls or contract-registry changes
• Deterministic capabilities: PASS - no capability derivation or resolver changes
• RBAC-UX: PASS - no /admin, /system, tenant context, workspace context, authorization, or capability behavior changes
• Workspace isolation: PASS - no workspace data or workspace-scoped route behavior changes
• RBAC-UX destructive-like actions: PASS - no destructive actions
• RBAC-UX global search: PASS - no Filament or global-search changes
• Tenant isolation: PASS - no tenant data, tenant reads, or tenant routes
• Run observability: PASS - no long-running, remote, queued, or scheduled product work
• OperationRun start UX: PASS - no OperationRun behavior
• Ops-UX 3-surface feedback: PASS - no OperationRun notifications or lifecycle output
• Ops-UX lifecycle: PASS - no OperationRun.status or OperationRun.outcome changes
• Ops-UX summary counts: PASS - no summary-count semantics
• Ops-UX guards: PASS - no Ops-UX guard changes
• Ops-UX system runs: PASS - no system-run behavior
• Automation: PASS - no queue, retry, lock, idempotency, or backoff behavior
• Data minimization: PASS - public static copy and metadata only; no secrets, tokens, or tenant data
• Test governance (TEST-GOV-001): PASS - browser/static classification is explicit, uses existing website lanes, and introduces no hidden Laravel/Filament/provider/database setup cost
• Proportionality (PROP-001): PASS - website-local narrative and metadata updates only; no new product structure or semantic machinery
• No premature abstraction (ABSTR-001): PASS - no new factories, registries, resolvers, strategies, interfaces, or pipelines
• Persisted truth (PERSIST-001): PASS - no new persisted product truth or artifacts beyond existing static build output
• Behavioral state (STATE-001): PASS - no new product states, statuses, or reason families
• UI semantics (UI-SEM-001): PASS - public copy and labels remain local presentation, not a shared semantic framework
• Shared pattern first (XCUT-001): PASS - existing shared website copy, layout, metadata, navigation, and smoke helpers are reused
• Provider boundary (PROV-001): PASS - public provider vocabulary is explicitly bounded to positioning only; no platform-core coupling is added
• V1 explicitness / few layers (V1-EXP-001, LAYER-001): PASS - direct website-local edits only
• Spec discipline / bloat check (SPEC-DISC-001, BLOAT-001): PASS - no enum, DTO, presenter, persisted entity, interface, registry, resolver, or taxonomy is introduced
• Badge semantics (BADGE-001): PASS - no shared badge/status taxonomy changes
• Filament-native UI (UI-FIL-001): PASS - no Filament UI
• UI/UX surface taxonomy: PASS - no operator-facing surface
• Decision-first operating model: PASS - public visitor decision flow is improved, but no operator decision surface is added
• Audience-aware disclosure: PASS - trust/proof boundaries are stated conservatively without exposing operator/raw evidence surfaces
• UI/UX inspect model: PASS - no operator list/detail surface
• UI/UX action hierarchy: PASS - no Filament actions or admin action surfaces
• UI/UX scope, truth, and naming: PASS - public category language, provider posture, and CTA vocabulary stay honest and non-implementation-first
• UI/UX placeholder ban: PASS - placeholder links and fake pages are explicitly banned by this feature
• UI naming: PASS - public CTA labels map to real next steps and avoid unsupported workflow verbs
• Operator surfaces: PASS - no /admin surface changes
• Filament UI Action Surface Contract: PASS - no Filament Resource/RelationManager/Page changes
• Filament UI UX-001: PASS - no Filament screen changes
• Action-surface discipline: PASS - no operator action surface changes
• UI review workflow: PASS - website-specific shared patterns and public validation responsibilities stay explicit without widening into platform scope

Initial Gate Result: PASS - no constitution violations or unresolved clarifications.

Test Governance Check

• Test purpose / classification by changed surface: Browser/static website
• Affected validation lanes: website build, public smoke, manual browser review, static claim scan, whitespace/scope checks
• Why this lane mix is the narrowest sufficient proof: The feature changes public copy, route metadata, CTA intent, navigation exposure, and claim discipline. Laravel/Pest/Filament lanes would not prove the changed behavior.
• Narrowest proving command(s): cd /Users/ahmeddarrazi/Documents/projects/wt-website && corepack pnpm build:website; cd /Users/ahmeddarrazi/Documents/projects/wt-website && WEBSITE_PORT=4321 corepack pnpm --filter @tenantatlas/website test:smoke; cd /Users/ahmeddarrazi/Documents/projects/wt-website && grep -RIn -e 'href="#"' -e 'Intune Management Tool' -e 'Intune backup tool' -e 'DSGVO compliant' -e 'GDPR compliant' -e 'ISO certified' -e 'Google supported' -e 'AWS supported' -e 'automatic restore' -e 'autonomous remediation' -e 'neutral SaaS visual' -e 'lorem ipsum' apps/website/src apps/website/public 2>/dev/null || true; cd /Users/ahmeddarrazi/Documents/projects/wt-website && git diff --check; cd /Users/ahmeddarrazi/Documents/projects/wt-website && git status --short -- apps/platform
• Fixture / helper / factory / seed / context cost risks: none - no database, provider, workspace, membership, session, queue, Sail, Laravel, Filament, or Livewire setup
• Expensive defaults or shared helper growth introduced?: no
• Heavy-family additions, promotions, or visibility changes: none - existing Playwright smoke remains explicit and website-local
• Surface-class relief / special coverage rule: N/A - public website
• Closing validation and reviewer handoff: Reviewers should rely on website build, Playwright smoke, static claim scan, desktop/mobile manual review, and apps/platform untouched confirmation. If copy changes add new public docs or navigation surfaces, smoke route allowlists and metadata expectations must be updated in the same feature.
• Budget / baseline / trend follow-up: none expected
• Review-stop questions: lane fit, claim drift, placeholder-link drift, route-exposure drift, hidden platform coupling
• Escalation path: document-in-feature
• Active feature PR close-out entry: Smoke Coverage
• Why no dedicated follow-up spec is needed: The validation cost remains local to this public website positioning pass unless future website work creates a recurring release-governance problem.

Project Structure

Documentation (this feature)

/Users/ahmeddarrazi/Documents/projects/wt-website/specs/404-public-content-messaging/
├── plan.md
├── research.md
├── data-model.md
├── quickstart.md
├── contracts/
│   └── public-content-contract.md
└── tasks.md

Source Code (repository root)

/Users/ahmeddarrazi/Documents/projects/wt-website/apps/website/
├── astro.config.mjs
├── package.json
├── playwright.config.ts
├── process-html.mjs
├── public/
├── src/
│   ├── components/
│   │   ├── pages/
│   │   └── sections/
│   ├── content/
│   │   ├── docs/
│   │   ├── blog/
│   │   ├── insights/
│   │   └── products/
│   ├── data_files/
│   ├── layouts/
│   ├── pages/
│   │   └── en/
│   └── utils/
└── tests/
    └── smoke/

Structure Decision: Use the existing /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website Astro application and its current localized route/component/content organization. Do not create new base folders and do not touch /Users/ahmeddarrazi/Documents/projects/wt-website/apps/platform.

Complexity Tracking

Violation	Why Needed	Simpler Alternative Rejected Because
None	N/A	N/A

Proportionality Review

• Current operator problem: Public evaluators and reviewers still receive the wrong product category and an incomplete governance narrative from the current website.
• Existing structure is insufficient because: The website foundation is already stable, but its public copy, metadata, navigation, and trust/provider boundaries do not yet express the intended policy-governance positioning.
• Narrowest correct implementation: Update the existing website-local copy system, page hierarchy, docs exposure, metadata, and smoke expectations inside apps/website only.
• Ownership cost created: Ongoing maintenance of public positioning copy, provider/trust claim guardrails, and smoke expectations for emitted public routes.
• Alternative intentionally rejected: A broad website redesign, a new content system, and any apps/platform-linked implementation or provider runtime work.
• Release truth: Current-release public website positioning truth.

Phase 0 Research

Research output is captured in /Users/ahmeddarrazi/Documents/projects/wt-website/specs/404-public-content-messaging/research.md.

Resolved clarifications:

• The active website remains the existing Astro 6 app in /Users/ahmeddarrazi/Documents/projects/wt-website/apps/website; no framework decision is needed.
• Core public routes are thin wrappers that delegate to shared page components in src/components/pages.
• The primary copy, navigation, CTA labels, and per-route metadata are centralized in src/data_files/site-copy.ts.
• German default routes and /en/... mirrors share the same content source through locale-keyed records rather than separate content systems.
• /product is a redirect alias to /platform, so the governance model should stay anchored to /platform and not a second product page.
• Existing Playwright smoke helpers already cover rendered routes, redirect aliases, placeholder-link bans, forbidden public residue, metadata, and mobile/keyboard/overflow checks.
• Public docs routes are intentionally emitted and must stay aligned with the same positioning/claim contract as the core marketing pages.
• No REST, GraphQL, database, queue, Laravel, Filament, Livewire, or provider-runtime contract is required for this feature.

Phase 1 Design

Design output is captured in:

• /Users/ahmeddarrazi/Documents/projects/wt-website/specs/404-public-content-messaging/data-model.md
• /Users/ahmeddarrazi/Documents/projects/wt-website/specs/404-public-content-messaging/contracts/public-content-contract.md
• /Users/ahmeddarrazi/Documents/projects/wt-website/specs/404-public-content-messaging/quickstart.md

The design treats public route behavior, messaging claims, provider posture, CTAs, operating-model sections, and route metadata as the contract. No REST, GraphQL, database, Laravel, Filament, Livewire, Microsoft Graph, queue, job, RBAC, or runtime platform contract is introduced.

Post-Design Constitution Check

Post-Design Gate Result: PASS

• Phase 1 remains website-local and scoped to apps/website.
• All clarification markers are resolved.
• No product persistence, abstraction, status family, provider runtime seam, OperationRun behavior, RBAC behavior, or Filament behavior is introduced.
• Shared-pattern reuse stays within the existing website copy/layout/metadata/smoke system.
• Provider vocabulary remains bounded to public positioning only.
• Validation remains explicit and limited to website build, smoke, claim scans, and scope checks.
• Agent context must be updated with the current plan outputs before implementation continues.