PDF X-Ray How it works

What is your PDF
hiding from you?

AI tools can read text that is completely invisible to the human eye — including secret instructions embedded by bad actors to manipulate how the AI analyzes your documents. This tool shows you everything. Learn how it works ↓

Your file is never uploaded — all analysis happens on your device
Drop your PDF here

or click to choose a file from your device

Supports any standard PDF. Scanned documents (images without text layer) cannot be analyzed.
View all extracted text

How it works

PDF X-Ray uses PDF.js to read the raw text layer of your document — the same data that AI assistants process when you share a PDF. It checks every text item against four known techniques used to hide content from human readers while keeping it fully machine-readable.

Tiny font size

Text rendered at font sizes below 3pt is effectively invisible to the naked eye but fully readable by AI. A document can embed entire paragraphs of hidden instructions at 0.1pt — too small to see, but present in the text stream.

Off-page positioning

Text placed at coordinates outside the visible page boundaries never appears on screen or in print, but stays in the document's content stream. AI assistants process it alongside the visible content.

White text

Text colored to match the page background is invisible unless selected with a cursor — a classic trick that fools human readers but not AI models, which read color-blind from the raw text layer.

Invisible rendering mode

The PDF specification includes a text rendering mode that hides text from being drawn on screen while keeping it in the content stream. The text is completely invisible in any viewer but still present for AI to read.