Privacy
How PDF Checkup handles the PDFs you check: processed on our checker server and deleted right after, never kept, never used to train AI. No accounts, no cookies, no third-party trackers.
Short version: the PDF you upload is sent to our checker server, checked (and fixed, if you ask), and deleted as soon as the result is sent back. We keep no copy and never use documents to train AI. There are no accounts, no cookies and no third-party trackers. If you buy a remediation pack, Stripe handles the payment and we keep a small order record without your email address.
The PDFs you upload
When you choose a PDF, your browser sends it over HTTPS to our checker service. The file is written to a temporary folder, checked with veraPDF and our own checks, and the folder is deleted when the response is sent, normally within a few seconds. If the service crashes mid-check, leftover files are wiped automatically within 10 minutes. We do not log file names or document text. The stated maximum retention for any uploaded file is one hour.
The report is shown in your browser and is not stored by us. Fixes and compliance reports work the same way: you send the file again, we return the fixed file or the report, and delete everything.
The checker service runs on Railway (United States). The website runs on Cloudflare Workers.
AI alt text
If you use AI alt text (paid packs only), small crops of the images you choose to describe and up to 700 characters of text from the same page are sent to a vision model on Cloudflare Workers AI. We keep nothing, and Cloudflare does not use Workers AI inputs to train models. You review every suggestion before it is written into your PDF.
What not to upload
Do not upload documents you are not allowed to share with a service provider, such as files covered by a confidentiality agreement, or medical, legal or financial records about other people unless you have a basis to process them.
Usage events
The page sends small anonymous events so we can see whether the tool is useful: that a page loaded, that a file was checked, fixed or reported, and which checkout option was clicked. Events never contain your file, its name or its content, and are not linked to a visitor id or cookie.
Purchases
Payments are handled by Stripe, which receives your email address, card details and billing country/ZIP under its own privacy policy. We never see or store your card.
For each purchase we store an order record: the order code, what you bought, credits used, the purchase and any refund time, the amount, Stripe's payment reference, and a one-way hash of your checkout email so you can restore the purchase. To charge each document once, we store a one-way hash (SHA-256) of each file you fix, linked to your order; it cannot be turned back into the file. We do not store the email address itself. Your browser keeps the unlock token in its local storage. Order records are kept while the purchase can be used and as long as accounting and tax rules require.
No cookies, no third-party tracking
The site sets no cookies and loads no third-party analytics, advertising or tracking scripts.
What our hosts see
Like any website, Cloudflare and Railway receive the technical data needed to deliver the service (such as IP address, user agent and URL) and keep short-lived request logs for security and reliability. See Cloudflare's privacy policy and Railway's privacy policy.
Changes
We will update this page before the way the site handles data changes. Last updated November 17, 2026.