Invoice extractor
Extract invoice fields from PDFs and photos into one checked table: numbers, dates, parties, line items, tax and totals — with the arithmetic verified locally and every field backed by source evidence.
About this tool
Bookkeeping starts with a pile of invoice PDFs and photos and a blank spreadsheet. This extractor closes that gap: it reads up to 20 invoices per batch — text-layer PDFs are parsed locally in your browser, photos and scans go to the vision model behind your own API key — and returns one row per invoice with the number, dates, parties, tax ids, subtotal, tax and total, plus a separate line-item table.
What makes it safe to use is the check that happens after extraction, on your device: the line items are re-added and compared with the subtotal, and subtotal plus tax is compared with the total. An invoice whose numbers do not reconcile is flagged with the reason instead of being silently entered into your books. Every row also carries a short verbatim quote from the invoice, so a human can spot-check any field in seconds.
How to use Invoice extractor
- Add your provider API key in the key panel (stored only in your browser).
- Drop up to 20 invoice PDFs or photos. Text-layer PDFs are read locally; photos go to the vision model you chose.
- Review the table: totals are re-added locally and any invoice whose numbers do not reconcile is flagged.
- Export the summary and line items as CSV or JSON.
Keep the original files and inspect the exported result before publishing or sharing it.
Frequently asked questions
How are the amounts verified?
After extraction, the tool re-adds the line items and checks them against the subtotal, and subtotal + tax against the total, on your device. Mismatches are flagged in the Check column — they are never silently passed.
Which files are supported?
PDF invoices (with a text layer) and JPG/PNG/WebP photos or scans. Text-less PDFs are routed to the vision model like photos.
Where does my invoice data go?
Extracted text or the image is sent only to the AI provider behind your own key, directly from your browser. Localcraft has no server in between.
How many invoices can I run at once?
Up to 20 per batch. Each invoice is one model call, so your provider usage scales with the batch.
Related tools
- AI chat — Chat with OpenAI, Claude or Gemini using your own API key. Your key stays in your browser; requests go straight to the provider.
- AI summarizer — Summarize long text with your own OpenAI, Claude or Gemini API key. Brief, detailed or bullet-point summaries, all in your browser.
- AI translator — Translate text between 30 languages with your own OpenAI, Claude or Gemini API key. Private by design: key and text stay with you and the provider.
- AI writing assistant — Improve, expand, shorten or re-tone your writing with your own API key. Six editing modes, direct browser-to-provider calls.
- AI code explainer — Paste code and get a plain-English explanation using your own OpenAI, Claude or Gemini API key. Language auto-detect included.
- AI PDF chat — Upload a PDF and ask questions about it. The PDF is parsed in your browser; only the extracted text and your question go to your chosen AI provider.
- AI image analyzer — Upload an image and let AI describe it, extract text (OCR), or identify objects. Uses vision models from OpenAI, Claude or Gemini with your own key.
- AI voice generator — Turn text into natural-sounding speech with OpenAI TTS. Six voices, speed control, play in the browser and download MP3.
- AI data analyst — Upload a CSV and ask questions in plain English. The file is parsed locally; the AI sees only a schema summary and sample rows.
- AI spend analyzer — Turn your AI provider usage export into an independent cost receipt: per-model and per-day breakdowns, spike alerts and a monthly projection. Runs 100% in your browser — your billing data is never uploaded.