Security & privacy — the specific version
Not a badge row. Exactly what happens to a statement, step by step.
Where your file goes
Nowhere. The PDF is opened and read by PDF.js inside your own browser. The file itself is never uploaded to SheetRecon or anyone else.
What is sent, and to whom
Only the extracted rows — date, description, amount, and the printed opening/closing balances — are sent over an encrypted (HTTPS) connection to the SheetRecon checker, a function hosted on Cloudflare (our sole sub-processor for the check). It computes the reconciliation verdict and returns it. It has no database and stores nothing.
Deletion
There is nothing to delete on our side: no file, no rows, and no statement content are retained after the verdict is returned. In your browser, the “Delete my data” control in the tool clears the extracted rows and the local conversion counter immediately.
Training
Your files and rows are never used to train any model. There is no model in the pipeline — extraction is text-layer parsing and the check is arithmetic.
Payments
Payments run on Stripe; SheetRecon never sees or stores card numbers. A data-processing agreement is available on request — email [email protected].
Honest limits
This release reads digital PDFs only. Browser-local metering (the free first conversion) lives in your browser's local storage and can be cleared with the Delete control — we would rather meter honestly on your device than demand an account to meter on ours.