Security

Nothing is kept unless you ask.

A statement is read in memory and dropped when the response is sent. This page lists everything we do store, how long we keep it, how a number becomes trustworthy, and what we do not claim.

What we store, and for how long

WhatHowHow long
The statement PDF and its rowsIn memory for the request onlyDropped when the response is sent
A PDF's passwordIn memory, to open the fileNever stored or logged
Async job resultsEncrypted1 hour after the job ends, then deleted
A statement kept so you can ask questions again (opt-in)EncryptedWhat you asked for (retain_minutes), at most 24 hours
Request logCounts, timings, bank name, error codes and model cost. No transactions, names or amountsUsed for your usage charts
Webhook delivery logEvent ids, status and counts. No statement data30 days
AccountEmail, password hashed with argon2, linked Google or GitHub idUntil you delete the account
API keysHash only. The key is shown onceUntil revoked
A file you attach to a bad-parse report (opt-in)Only if you choose to attach itUntil the parser is fixed

Pages that need a language model are sent to our model provider, Anthropic, under its API data terms. Pages read from the layout, and scans read by free OCR, never leave our server. An account or a single API key can switch AI off completely.

How a number becomes trustworthy

  • Every balance is re-checked

    Each row's balance must equal the previous balance plus credit minus debit, to the kobo, oldest first or newest first. Breaks are listed with their page and row.

  • Model readings are matched against the page

    When a language model reads a page, every value it returns is matched word for word against the page's own text. A number that is not printed on the page is caught, not trusted.

  • Repairs are reported, never silent

    Months stapled into one PDF are split into sections, each balanced on its own. Rows a bank printed twice are removed and listed. A brought-forward line is not counted as money in.

  • A grade your code can act on

    Every response carries quality high, medium or low with plain-word reasons. Auto-accept high; route the rest to a person.

How we protect the service and your customers

  • Questions in English never show the statement to the AI

    The model turns your question into a query. Our code runs the query on the checked rows, and you get the data plus the exact query, so the answer can be repeated without the model.

  • AI can be switched off

    An account or a single key can turn AI off. Every page is then read on our servers only, scans included, through free OCR, and English questions are off. GraphQL still works.

  • Webhooks are signed and cannot be aimed inward

    Events are signed with HMAC-SHA256 and a timestamp, carry ids and counts but never statement data, and are only sent to https addresses that resolve to public IPs, checked again before every send.

  • Keys and access

    Live keys are hashed at rest and revocable, and can be limited to certain IP addresses or an expiry date. Sessions are httpOnly cookies with an origin check. Sign-up, login, reset and parsing are rate limited.

  • Heavy work is isolated

    Every parse runs in a separate worker process, so a huge or broken file fails only its own request and your other calls stay fast.

What we do not claim

We do not hold SOC 2 or ISO certifications today and we do not publish an uptime figure yet. You must have your end user's consent to process their statement.

Read the Privacy Policy and Terms of Service for the legal wording. To report a security problem or ask a question, email support@sumproof.dev. These facts are also available as markdown.