LT Google Review Booster

Inspiration

Local businesses need a practical way to invite genuine customers to share feedback without rating gates, incentives, or pressure. A review-request tool should make follow-up consistent while protecting each customer's freedom to leave an honest review—or no review at all.

What it does

LT Google Review Booster is a compliance-first Work & Productivity workflow for neutral Google-review requests. Merchants can prepare on-screen QR, printed QR, short-link, receipt, invoice, NFC, and future completed-service placements. Each customer-facing path keeps the Google-review and private-feedback actions independent: no rating question, no sentiment routing, no reward, and no requirement to post while staff watch.

Working demo

A free, credential-free synthetic demo is available at: https://lt-google-review-prod-dashboard-web.vercel.app/demo

It shows a deterministic merchant workflow: choose a synthetic location, activate or pause an on-screen QR placement, open a separate customer route, view two neutral customer choices, and inspect non-persisted event evidence. The QR encodes only a synthetic LT route. The "Review on Google" action intentionally fails closed because this contest demo has no live Google destination; private feedback is acknowledged without collecting text or customer data. The demo uses synthetic data only and needs no login, payment, API key, or judge account.

How it is built

I am a solo AI-assisted developer leading the product, compliance, architecture, implementation, and release decisions. I used OpenAI Codex and GPT-5.6 throughout the build. Sol is the main coordination role; Terra, Luna, and Spark are bounded subagents dispatched in the same Codex task under exact non-overlapping file leases. These are coordination labels, not claims about separate or concurrent engines.

The project is a Node.js 24 and TypeScript monorepo with Next.js dashboard and funnel surfaces, an Android/Expo shell, shared policy/domain/contract packages, Supabase migration contracts, regional routing boundaries, security headers, secret scanning, a CycloneDX SBOM contract, and automated tests.

What works now

• A hosted, interactive synthetic merchant-to-customer workflow.
• Active, paused, resumed, and fail-closed on-screen QR placement states.
• A standards-based QR carrying a fixed synthetic opaque-shaped route.
• Independent neutral Google and private-feedback customer choices.
• A source-derived pricing preview, request-channel inventory, capability readiness decisions, campaign lifecycle, launch preparation, delivery lifecycle, and contest overview.
• Root checks covering contracts, migration audits, lint, typecheck, automated tests, Next.js builds, fixtures, and secret scanning.

Challenge and impact

The challenge is making a review-request workflow useful without enabling manipulation. The product encodes compliance in workflow states and tests: it rejects rating gates, incentives, fabricated reviews, sentiment routing, and unverified provider or cross-region paths. The intended audience is owner-operated and multi-location local businesses whose receptionist, cashier, or service teams need a consistent post-completion invitation without customer pressure.

What I learned

Compliance works best when it is executable. Neutral copy, equal action visibility, consent evidence, URL restrictions, regional isolation, and fail-closed provider controls belong in product contracts and tests—not only in a policy document.

Current scope

This is a runnable contest vertical slice, not a production release. Live Google destination resolution, messaging, billing, production authentication, persistent audit storage, and Google-review posting are intentionally disabled pending their provider, privacy, and regional controls. The video and description claim only the working hosted synthetic demo shown above.

Built With

• cloudflare
• codex
• expo.io
• github
• gpt-5.6
• next.js
• node.js
• openai
• supabase
• typescript
• vercel