How to Check a PDF for Hidden Text

A PDF can carry text you never see. The page looks clean, but the file underneath holds words painted the same color as the background, shrunk to a fraction of a point, pushed past the edge of the page, or tucked into a layer that is switched off. None of it shows when you read the document. All of it is still there, and any program that extracts the text — a search index, a résumé parser, an AI summarizer — reads it in full.

This guide covers two ways to check: fast manual methods you can do in any PDF viewer, and a structural scan that catches the cases the manual methods miss.

Why hidden text exists at all

To understand where text hides, it helps to know how a PDF stores it. A PDF keeps two things that feel like they should be the same but are not: the visual page, which is a set of drawing instructions, and the text layer, which is the actual character data placed at specific coordinates. When you select text with your cursor, you are selecting from the text layer. When you look at the page, you are seeing the rendered result of the drawing instructions.

Those two can disagree. Text can sit in the layer while the drawing instructions make it invisible — same color as the page, zero-size, off to the side, or explicitly set to a “render but paint nothing” mode. The words are fully present for extraction and fully absent for a human. That gap is the whole game.

The fastest manual test: select all and copy

The quickest check needs no tools. Open the PDF, press Ctrl+A (Cmd+A on a Mac) to select everything on the page, then copy it and paste into a plain-text editor.

Now compare. If the pasted text contains sentences, paragraphs, or instructions that were not visible on the page, you have found hidden text. This works because “select all” grabs the entire text layer regardless of how it is drawn. White-on-white text highlights and copies just like normal text — the selection box will often reveal it as a faint colored band over what looked like empty space.

Watch for two tells while you drag-select:

• A highlight rectangle over blank areas of the page. If empty margins or gaps light up when you select, something is there.
• Pasted content longer than what you read. A one-page document that pastes three pages of text is hiding two pages somewhere.

This method is reliable for the most common trick — text colored to match the background — and it costs ten seconds.

Its limits

Select-and-copy is a good first pass, not a complete one. It misses several real cases:

• Off-page text that sits outside the crop area may not be selected by a normal drag, and may not be included when you copy the visible page.
• Hidden layers (optional content that is toggled off) usually do not render or select, yet the text stays in the file and remains extractable by other software.
• Tiny fonts can be so small you never notice the highlight, especially against a busy background.
• Annotations — comments and sticky-note text — live in a separate structure and often will not come along with a page copy at all.

In other words, the manual test tells you when something is obviously wrong. It cannot tell you the file is clean.

Turning on hidden layers manually

If you want to chase layers by hand, some desktop viewers expose them. In Adobe Acrobat, the Layers panel in the sidebar lists any optional-content groups and lets you toggle their visibility. Switch every layer on and re-read the page. Content that appears only when you enable a normally-off layer is content the author chose not to show you by default. Most viewers, including browser PDF viewers, give you no layers panel at all, which is exactly why layer-based hiding works.

The structural scan

The thorough way to check a PDF is to read its structure directly instead of trusting what any viewer chooses to display. A structural scan walks the text layer run by run and inspects the properties the eye cannot judge: the exact fill color against the page background, the point size, the coordinates relative to the crop box, the render mode, the layer each run belongs to, plus the parts of the file that never render at all — embedded attachments, JavaScript, and annotation contents.

That is what this site’s scanner does. You drop a PDF in, and it decodes the file and reports every run of text that is invisible or near-invisible, every layer that is hidden, and any text matching known instructions aimed at AI readers. It reads properties, not pixels, so white-on-white, render-mode-3, and tiny-font text all surface the same way — with the actual hidden words quoted and the page number listed, so you can judge each finding yourself. Text cropped fully off the page is the one exception: the PDF engine clips it before extraction ever sees it, so there is nothing to quote. What the scan reports instead is the crop box mismatch that goes along with it — the page’s crop area is smaller than its full media size, a sign that something was trimmed out of view.

It runs entirely in your browser. The file is decoded locally in a WebAssembly sandbox and never uploaded, which matters when the document you are checking is a contract, a résumé you received, or an unpublished paper you are reviewing.

A practical order of operations

1. Select all, copy, paste into a text editor. Read what comes out and compare it to the page.
2. If your viewer has a Layers panel, turn every layer on and re-read.
3. Run a structural scan to catch crop box mismatches (the sign left behind when content is cropped off the page), hidden layers, tiny fonts, render-mode tricks, and injection patterns the manual pass cannot see.
4. Read the flagged text in context. Not every finding is malicious — a trimmed scan or a legitimate figure callout can trip a detector — but every finding is text someone put in the file and kept off the page.

No single check is complete on its own, and even a clean structural scan does not prove a file is safe: text baked into an image, for instance, is invisible to text-layer analysis. But the combination of a copy-paste test and a structural scan will catch the overwhelming majority of hidden-text tricks in circulation, and it takes a minute.

Want to check a file right now? Scan a PDF for hidden text with the tool on the homepage. It runs locally, and nothing leaves your browser.