PDF accessibility preflight: check title, language, tags, and alt text

This is a heuristic preflight, not a certification. It looks for a few common signals. It is not a PDF/UA validator, does not certify accessibility, and is not legal advice. A result with no issues does not mean the PDF is fully accessible or meets any law or standard.
Some things cannot be decided automatically. When a check cannot tell, the result says “Could not be determined” and suggests what a person should look at. Reading order is never confirmed as correct, and the quality of alternative text is not judged.
Test with people and assistive technology. For a full review, use a dedicated accessibility checker, read the document with a screen reader, and ask people who use assistive technology.
Nothing in the PDF is changed. This tool only reads the file and shows results.

Document title: whether a title is set in the PDF’s properties and whether viewers are told to show it. The file name does not count as a title.
Document language: whether a language code such as “en-US” is set.
Tags: whether the PDF is marked as tagged and has a structure tree.
Alternative text: whether tagged figures have a text description, and whether images appear in a PDF without tags.
Selectable text: whether each page has text that a screen reader can read, or looks scanned.
Reading order: signals that the order may be a problem, such as missing tags or text stored out of visual order.

Choose a PDF from your device, or drop it on the box above.
Press Start check. Nothing happens until you press the button.
Read each result. “Issue found” points to a likely problem, “Could not be determined” means a person needs to check, and “No issue found” only means this check did not find a problem.

Everything happens in this browser tab. The PDF, its text, and the results are not sent anywhere.

Use the tool