Coming SoonWindows

Turn a pile of scans into files you can file.

Paperwork photographed “for later” is still unnamed later. 2cmd™ Ledger is Coming Soon: local reading of supported scans and PDFs, conservative name suggestions, and a preview before anything is renamed or moved.

Scans keep the page and lose the identity

A Saturday with a scanner produces scan002.pdf through scan048.pdf. Some pages are receipts, some are contracts, some are photos of a whiteboard. The files are images of useful paper with useless names.

Cloud “scan to Drive” workflows upload the page to name it. That is the wrong default when the stack includes tax documents, IDs adjacent to receipts, or client paperwork.

How Ledger would work with scans

Choose the folder of scans. Ledger is intended to read digital PDFs and to OCR images and scans where needed, for PDF, JPG, PNG, and WEBP.

Supported v1 classes are Receipt, Invoice, Bank Statement, and Other. Photographed documents follow the document class; Photo is not a separate class. Contracts are outside initial v1 intelligence. A scan it cannot classify with confidence should land in Other.

You would preview, edit, and apply. OCR mistakes are expected; the product is built around review, not around trusting every suggested field.

Local processing

Scans are often the documents people are least willing to upload. Ledger’s intended v1 processes them on the Windows PC. No customer document upload is required for that design.

Nothing changes until you approve it

Suggestions are not applied in the background. You confirm a batch. Collision safeguards and undo of the last transaction are part of the safety model.

What this is not

This is not a photo manager, a general OCR suite, or a promise of verified multilingual accuracy. Video, DOCX, and XLSX are out of v1. Broad photo-library organization is an explicit exclusion.