Skip to content

heymegabyte/agent-skills / 06-build-and-slice-loop

build-and-slice-loop

Verified

Use when implementing features or streaming OnePrompt site builds. Complete vertical slices with real content, strict TypeScript and Zod. Parallel section/data work, deterministic assembly, granular patches and immediate valid previews.

06 — Build and Slice Loop

Implement features in real vertical slices (UI→API→DB→tests→deploy) starting with the homepage; no placeholders, no stubs.

Vertical-slice principle

Complete each feature end-to-end through its required layers (UI → API → DB → tests → deploy). Independent slices may proceed in parallel; the slice contract defines completeness, not a global serial execution requirement.

Homepage FIRST (every project, no exceptions)

  • First slice is always the homepage hero + nav + footer
  • Real H1, real meta tags, real OG card, real JSON-LD
  • Deploy before adding any other route — establishes design system, auth layer, analytics tier, deploy pipeline
  • Per rules/website-build-doctrine.md Phase 0 / Phase 1

For the OnePrompt streaming builder, render the cached draftl.ink/Storybook shell immediately while independent research/section/data work proceeds. Content-only slot patches do not rebuild the project. Follow rules/event-sourced-build-progress.md: typed agent results, one assembler, deterministic scripts/ast-grep, versioned patches and an isolated preview iframe. Preserve normal validation/promotion gates; a draft preview is not a production deployment.

Anti-placeholder (NON-NEGOTIABLE)

  • ❌ Lorem ipsum
  • ❌ TODO / FIXME / TBD in shipped user-visible strings (source-comment TODOs OK per rules/todos-are-roadmap.md)
  • ❌ Gray placeholder boxes / silhouettes
  • ❌ "Coming soon" without firm date
  • ❌ Stub images / generic SVGs
  • ❌ "John Doe" / "Jane Doe" / "company.com" / "example.com"
  • ❌ Bracket placeholders [Insert Name] / [Your Title]
  • ❌ Single-character "x" / "?" labels

Per rules/copy-writing.md § Production-review copy gate. Build validator greps dist/ for these patterns.

Slice contract (every slice ships ALL of these)

  1. Types + Zod schemas at boundaries (rules/zod-everywhere.md)
  2. Contract-first API (typed request/response)
  3. Loading + empty + error + success states (4-state system per 10-experience-and-design-system)
  4. Accessibility (axe 0 violations + WCAG 2.2 AA per _kernel/standards.md#wcag22)
  5. 6-breakpoint responsive (per _kernel/standards.md#breakpoints)
  6. Observability (Sentry breadcrumb + PostHog event + Workers Trace span)
  7. Tests (Vitest unit + Playwright E2E from homepage outward per rules/e2e-testing.md)
  8. Feature flag where rollout risk exists (rules/feature-flags.md)
  9. Tenant isolation (org_id on every row + every query, 404 on mismatch)
  10. Docs (JSDoc on exports + module README)

File organization

Worker (Hono)

src/worker/
├── index.ts              # entry — Sentry wrapper, route mounts, onError, notFound
├── db/
│   ├── schema.ts         # Drizzle source of truth
│   └── index.ts          # getDb helper
├── routes/<feature>.ts   # Hono sub-app per feature
├── lib/<feature>.ts      # business logic
├── middleware/           # auth, tenant, rate-limit
└── types.ts              # Env + Variables

Frontend (React + Vite)

Vite 8 (stable Mar 2026) builds on the unified Rolldown bundler — replaces esbuild+Rollup, 10–30× faster prod builds.

src/web/
├── main.tsx              # entry — router, providers
├── pages/<Page>.tsx      # one per route
├── components/
│   ├── ui/               # shadcn primitives
│   ├── <feature>/        # feature-specific components
│   └── shared/           # Header, Footer, ErrorBoundary
├── lib/                  # api client, auth context, hooks
└── styles.css            # Tailwind + design tokens

Shared

src/shared/
├── plans.ts              # pricing tiers
├── design-styles.ts      # design tokens shared web + worker
└── pageLayout.ts         # shared layout helpers

Tests

e2e/
├── FEATURES.md           # row-per-feature matrix (CI gate)
├── _fixtures/            # page objects
├── _helpers/             # snapshot, axe, breadcrumbs, visual
├── <feature>/
│   ├── happy-path.spec.ts
│   ├── edge-cases.spec.ts
│   ├── a11y.spec.ts
│   └── visual.spec.ts
└── _smoke/               # cross-feature smoke

Build loop (per slice)

  1. Write failing test (Playwright TDD-RED first per rules/e2e-testing.md)
  2. Author migration (drizzle/0NNN_<feature>.sql) + update db/schema.ts
  3. Apply locally (npm run db:apply:local)
  4. Implement (route, lib, page, components)
  5. Typecheck (npm run typecheck clean)
  6. Unit tests (npm test green)
  7. Run dev (npm run dev) — exercise in browser
  8. Deploy (npm run deploy)
  9. Prod E2E (npm run e2e:prod against PROD_URL)
  10. Verify console (zero errors, zero CSP violations, zero failed resources)
  11. AI vision QA (6bp screenshots → vision rubric ≥9/10)
  12. Commit + push (conventional commits per rules/main-only-branch.md)
  13. Update e2e/FEATURES.md + CHANGELOG

Real content sourcing

  • Copy: brand voice per 09-brand-and-content-system; never AI-default phrasing
  • Images: real photos > AI-generated brand-aligned > zero (skill 12)
  • Data: real API / Workers AI / D1 query > mocked
  • Testimonials: real attribution > none (never fabricated)
  • Stats: cited per rules/citations.md

CLI and compute work

Use native Go and the appropriate current Charm components per rules/terminal-styling.md. Keep product orchestration in TypeScript/Cloudflare; use Rust/WASM only when 05-architecture-and-stack/SKILL.md's measurement and runtime-fit gates pass. Do not serialize independent work for an animation or start a native runner for a content-only preview patch.

When to stop adding sections

Per rules/website-build-doctrine.md Phase 2: keep adding until next would dilute. Stop when "next hurts," never "we have enough."

See submodules: vertical-slice.md, file-organization.md, anti-placeholder.md.

receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementationStars: 321kUpdated Claude CodeCodexVerifiedInstall
requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirementsStars: 321kUpdated Claude CodeCodexVerifiedInstall
using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallbackStars: 321kUpdated Claude CodeCodexVerifiedInstall
verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions alwaysStars: 321kUpdated Claude CodeCodexVerifiedInstall