Shoebox: receipts in, spreadsheet out, and your clients’ data never leaves the laptop

Description and use

Shoebox lets freelancers and small shop owners quickly convert receipt photos into clean spreadsheets, keeping all data offline.

Problem

Freelancers and small shop owners spend hours manually entering receipt data into spreadsheets, while cloud OCR solutions expose sensitive client information to external servers.

Outcome

Shoebox reduces manual entry time by an estimated 7.5 hours per month for 300 receipts, eliminates cloud OCR costs, and protects client privacy by keeping all data offline.

Workflow

• 1. Install Ollama and pull Gemma 3:4b
• 2. Install project dependencies via pip
• 3. Run the local web app
• 4. Drag and drop receipt images (JPG, PNG, WEBP, HEIC)
• 5. The model extracts fields and displays a table
• 6. Review amber‑flagged rows for low confidence or math mismatches
• 7. Edit fields in the side drawer if needed
• 8. Export the table as a UTF‑8 BOM CSV

Evidence

Shoebox is a local web app. You drag in receipt photos (JPG, PNG, WEBP, HEIC). Gemma 3 (4B), running on your own machine through Ollama, reads each one and fills a table you can check, edit and export as CSV.

Source

For each receipt it extracts the date, vendor, currency, subtotal, tax/GST, total, payment method, a category, line items, and its own confidence score.

Source

It doesn't hide uncertainty. A row turns amber when confidence is under 0.6, fields are missing, or the line items differ from the total by more than 2%.

Source

The video walks through Shoebox from start to finish: the privacy panel, dropping in four receipt images and watching the scanner read them one at a time, the amber flags on rows the model isn't sure about, the side drawer for checking a row against its image, and the CSV export.

Source

Setup is ollama pull gemma3:4b, pip install -r requirements.txt, then run.bat (or run.sh).

Source