PDF/UA Auto-Tagger
Add the tags screen readers need to untagged PDFs.
PDF/UA-1 structure tags · veraPDF check before and after
What it does
Most PDFs made by software have no tags, so screen readers can't tell a heading from a footnote or a table from a paragraph, and accessibility checkers mark them as failing. Give this tool links to your PDFs and it adds the headings, paragraphs, lists, tables, links, language and title those readers need. In our test of 40 real-world PDFs, 39 looked exactly the same afterwards. It runs the open-source veraPDF checker on each file before and after, so you get the tagged PDF and a plain report of what now passes and what still needs a person.
What you get
- A tagged copy of each PDF, with a report you can check before you publish it.
- A before/after report per file from the open-source veraPDF checker, listing every rule that failed, what was fixed and what's left.
- Figures listed page by page, so you know exactly where alt text is needed. Supply it in the run and it goes into the file.
- Files that are already tagged keep their tags: the tool only finishes the document-level metadata (PDF/UA identifier, title, language if missing, display settings) and lists every change.
Best for
- PDFs made by software without tags: web-to-PDF tools such as WeasyPrint, LaTeX, LibreOffice with tagging off, and office print drivers.
- Working through a backlog of untagged reports, statements and manuals, with a person checking the results.
Not for
- Scanned PDFs (they need OCR first), fillable forms or password-protected files.
- Repairing existing tags. Current Word "Save as PDF" and Chrome "Print to PDF" output is usually tagged already; the tool finishes its document-level metadata but does not check or repair the tags themselves (reading order, headings, tables, alt text). It never invents a title: a file without one is reported as needing one.
- Untagged Word output and complex office layouts, for now. Word text is often split into many small paragraphs with too many headings, and tables and nested lists may come out as plain paragraphs. Check those files before you publish them.
- A compliance guarantee. Passing veraPDF covers the machine-checkable rules; reading order, heading levels and image descriptions still need a person.
Cost and how to start
- Open it on Apify Store and sign in, or create a free Apify account.
- Give it links to your PDFs and start the run.
- Download the tagged PDFs and the before/after report.
US$0.02 per page tagged. PDFs that were already tagged and only finished or left unchanged, and refused, failed or skipped files, aren't charged. Each run also has a start charge of US$0.00005 per GB of run memory. There's no separate charge for Apify platform usage.
Questions
- Does passing mean we're compliant?
- No. It means the file passes the automated PDF/UA-1 checks, the part a machine can verify. Laws such as the ADA and the EAA also need a human review of reading order and alt text, and the report lists what to look at.
- Which of my PDFs need this?
- Anything the report marks as untagged. As a rule of thumb, output from WeasyPrint, LaTeX, older Word versions and print-to-PDF drivers is usually untagged; current Word and Chrome output usually isn't.
- Where do my files go?
- They're downloaded into your own Apify run, processed there and stored only in your Apify storage.
Common problems
How we tested it
- A real 46-page standard. RFC 9562, published by the IETF, failed 12 veraPDF PDF/UA-1 rules as released. After one run on Apify it failed none.
- Checked independently. Every result is measured with veraPDF, the open-source validator from the Open Preservation Foundation and the PDF Association, not with a checker written for this tool.