PDF Inspector — what's actually inside a PDF (and how to read what we surface)
Most PDF tools tell you a single number. PDF Inspector tells you everything we can extract from a single parse of the document: page count, file size, page dimensions per page, mixed-orientation warnings, embedded fonts, attached files, hyperlinks, JavaScript handlers, form fields, encryption status, metadata fields. The point is not to overwhelm — it is that one parse already produces this data, and surfacing it once saves you running five different tools later. Here is what each field means, why each one matters in some real workflow, and the three audit patterns that use PDF Inspector as their starting point.
What a single parse surfaces
When you open a PDF in any modern reader, the reader's first job is to parse the file's structure — the page tree, the cross-reference table, the document catalog, embedded resources. That parse is fast (typically under a second for documents up to a few hundred pages) and produces a structured view of everything the PDF contains. PDF Inspector exposes that structured view directly.
The output covers six major categories: document-level metadata (size, page count, encryption), per-page properties (dimensions, orientation, content type), embedded resources (fonts, images, attachments), interactive elements (form fields, hyperlinks, annotations, JavaScript), accessibility signals (tag tree presence, language declaration), and creator metadata (author, title, subject, creation/modification dates).
Six categories of surfaced data and why each matters
Each section of the Inspector output serves a different workflow:
- Document overview. Page count, file size, version, encryption status, mixed-page-size warning. The 5-second triage before deciding what to do with the file.
- Page-by-page properties. Each page's width × height, orientation, rotation, and content type (text-based vs scanned). Critical for print prep and OCR decisions.
- Embedded fonts. Every font referenced in the document, whether it is embedded or relies on the reader having it installed. Non-embedded fonts are the #1 reason a PDF reflows on someone else's machine.
- Attached files. Any non-page content embedded in the PDF — XML invoice data, supplementary spreadsheets, exhibit photos. Often invisible in standard readers but indexed by our viewer.
- Interactive elements. Form fields, hyperlinks, annotations, named destinations, JavaScript handlers. The interactive layer most readers don't surface in one place.
- Accessibility signals. Tag tree presence (required for screen-reader access), language declaration, alt-text coverage on images. Helpful for compliance with WCAG, PDF/UA, Section 508.
Three audit patterns that start with Inspector
Where Inspector earns its place in a workflow:
- Pre-distribution audit. Before sending a PDF externally, run it through Inspector to verify: (a) fonts are embedded (otherwise the recipient sees different rendering); (b) metadata does not leak sensitive info; (c) JavaScript handlers do not raise red flags; (d) encryption is set correctly for the audience; (e) attached files are intended for inclusion.
- Forensic inspection of received PDFs. When evaluating a PDF you received — bid response, vendor contract, regulatory filing — Inspector surfaces the document's structural fingerprint: who created it, when, with what software, what version of PDF, what fonts. Useful in due-diligence and disputes.
- Pre-OCR / pre-conversion check. Before running OCR or PDF-to-text on a document, check whether it already has a text layer. Inspector reports text content per page; pages showing 0 extractable text are scans needing OCR.
Specific signals worth understanding
Fields in the Inspector output that have outsized importance in specific workflows:
- JavaScript handlers. Any handler is potentially executable code. Handlers that touch network, file system, or external apps are flagged high-severity. Review carefully before opening untrusted PDFs.
- Encryption: user-password vs owner-password. User-password (open-password) PDFs are fully encrypted. Owner-password (permissions-password) PDFs are readable but restricted. Inspector reports which kind, so you know whether you need a password to read it or whether you need our Unlock tool to remove restrictions.
- Mixed page sizes. A document that mixes Letter and A4 pages will render inconsistently across viewers and print awkwardly. Inspector flags this so you can normalize with Resize Pages.
- Embedded vs non-embedded fonts. Non-embedded fonts mean the reader has to find a matching font on the recipient's machine, and the match may differ. For print-ready documents, every font should be embedded. Inspector lists exactly which ones aren't.
- Mismatch between /Info dictionary and XMP stream. Both carry document metadata. When they disagree (different authors, different titles), the file was probably edited in a tool that updated only one. Inspector surfaces the mismatch as a warning.
How to act on the findings
Inspector is a read-only tool — it does not modify the PDF. Each finding suggests a follow-up:
- Non-embedded fonts → use Compress with Embed-Fonts option. Or regenerate the PDF from source with font embedding enabled.
- Sensitive metadata → Remove Metadata. Strip /Info and XMP fields before redistributing.
- Mixed page sizes → Resize Pages. Normalize to a single size for consistent printing and viewing.
- Hidden attachments → Extract Attachments. Pull the embedded files out and decide whether to keep or strip them.
- Tag-tree missing → Accessibility Checker. Run a full accessibility audit to see what's needed for compliance.
- JavaScript present → JS Detector. Get a per-handler view of what the code does so you can decide whether to keep or strip.
Limits and compatibility
On the free web tool, PDF Inspector handles PDFs up to 100 MB. Parsing runs in your browser via PDFium WebAssembly; nothing is uploaded. The Inspector output is exportable as JSON for downstream automation or as a readable report for inclusion in audit documentation.
Common pairings: Inspector → any modification tool, depending on what you find. Inspector is the cross-tool decision-maker — run it first when you don't yet know what the right next operation is.