Scanvice

AI receipt and invoice scanner.
Upload a receipt image and get structured, exportable data back — powered by Claude vision.

---

Scanvice is a landing site plus an AI-powered receipt and invoice scanner. Upload a photo of a receipt, invoice or bill and the app calls Anthropic's Claude vision model to read it and return clean, structured JSON — vendor, transaction details, totals, line items, a category guess and a per-field confidence score — which is rendered in the UI and can be exported (e.g. to XLSX).

Features

• Receipt/invoice extraction. POST /api/extract-receipt sends the uploaded image to Claude Sonnet 4.5 (vision) and returns structured data matching a typed ExtractedReceipt schema.
• Rich structured output. Vendor info, transaction details, totals and an amount breakdown, itemised line items, a category guess and per-field confidence scores; low-quality images still extract with lowered confidence and warnings.
• Rendered results. Vendor card, transaction details, line-items table, amount breakdown, confidence indicators and warning banners.
• Export. One-click export of results, including XLSX (via SheetJS).
• Server-side guardrails. Request-level rate limiting, image MIME-type and size validation (10 MB max), a request timeout, and metadata-only logging — receipt content is not retained.
• Marketing site + blog. Landing sections (hero, problem/solution, trust, CTA) and an MDX-style blog, with SEO built in (sitemap.ts, robots.ts, dynamic OG icon).

Tech stack

• Next.js 16 (App Router) with React 19 and TypeScript
• Tailwind CSS v4 with shadcn/ui + Radix primitives
• Anthropic Claude (claude-sonnet-4-5) vision via the Messages API
• SheetJS (xlsx) for spreadsheet export, Recharts, Framer Motion, Vercel Analytics

Getting started

Requirements

• Node.js 18.18+ (Next.js 16)
• An Anthropic API key

1. Install

npm install

2. Configure environment

Create a .env.local file in the project root:

ANTHROPIC_API_KEY=sk-ant-...

ANTHROPIC_API_KEY is required — the extraction endpoint returns a server_misconfigured error without it.

3. Run

npm run dev      # http://localhost:3000

Build

npm run build
npm run start
npm run lint

How extraction works

app/api/extract-receipt/route.ts runs on the Node.js runtime. It validates the upload (accepted image MIME types, 10 MB limit), applies a per-IP rate limit, and calls the Anthropic Messages API with the image and a strict extraction prompt (lib/receipt.ts). The model returns JSON only, which is parsed into the ExtractedReceipt type and sent back to the client for rendering and export.

License

Released under the MIT License © 2026 Olivier Lüthy. You're free to use, modify and distribute this software, including commercially, as long as the copyright notice and license are included.

Author

Built by Olivier Lüthy — GitHub.