Why PDFeyes exists

Most PDF tools treat every blank on a page as its own separate thing. That's fine for a one-page form. It falls apart the moment a document. Or a whole packet of them, repeats the same handful of facts dozens of times, which is exactly what real closing packets, onboarding bundles, and case files look like.

PDFeyes started from a simple observation: the people who deal with this most (real estate transaction coordinators, notary signing agents, HR teams, bookkeepers, small e-commerce sellers) aren't retyping information because they enjoy it. They're doing it because most tools make them.

The tools here look different on the surface but solve the same shape of problem: information is trapped in a PDF, or needs to get into one, and the manual route is slow and error-prone in ways that are hard to catch.

What it does today

Twenty tools, all running entirely in your browser. Nothing you load is ever uploaded to a server.

Fill a PDF

Fills any PDF form, including scans and flat forms with nothing to click. Type an answer once and place it wherever the form asks for it; correct it in one place and every copy changes. Signatures, initials, dates and check marks are placed the same way.

Statement → CSV

Turns a PDF bank statement into a spreadsheet. Rather than maintaining a template per bank. Which breaks silently whenever a bank redesigns its statement, it infers the layout from the document: finding date-led rows, clustering numbers into columns by position, and identifying a running balance by checking whether its changes are explained by the other columns. Where a balance exists, every row is checked against it, so the tool can verify its own output against arithmetic the bank already did.

Label Sheets

Paste a spreadsheet, get print-aligned labels on Avery-compatible sheets. Every layout's margins, label sizes, and gaps are checked to add up to the sheet dimensions exactly, and the rendered output has been measured against published specifications rather than assumed correct.

What it doesn't do yet

No OCR, which is the significant one: scanned or photographed documents have no text layer, so neither the PDF filler nor the statement converter can read them. There's also no AI-assisted detection for unusual placeholder styles, no saved profile vault, no batch mode across a folder of documents, and no image support on labels.

Each tool states its own limits plainly in the interface rather than burying them.

Where the name comes from

PDFeyes, the tool sees what's in your documents, and every place your information needs to go, so you don't have to go looking for it yourself.

On privacy

Every tool here runs client-side. That isn't a marketing position so much as a consequence of the work being computation that doesn't need a server. And it matters given what people put through these tools: closing packets full of personal data, bank statements, customer address lists. You can verify the claim rather than trust it: load any tool, disconnect from the internet, and it still works.