Is it safe to open a PDF? The real risks, and what a viewer should block
"Do not open attachments from people you do not know" is decent advice and not very specific. A PDF can do more than display pages, and the protections that matter are concrete and checkable.
What a PDF can contain beyond text and images
- JavaScript. The PDF format includes an embedded JavaScript engine (Acrobat JS). It was designed for interactive forms — calculating totals, validating fields — and it is a scripting language with file-system and network reach in some implementations.
- Embedded files. A PDF can carry arbitrary attachments inside itself, which is how malware arrives wrapped in a document that looks like an invoice.
- Launch and submit actions. Actions triggered by opening the document, clicking a link, or closing a page — including launching an application or submitting form data to a URL.
- Remote content and beacons. A remote image or a form submission tells the sender that you opened the file, when, and — with a unique URL per recipient — exactly who you are. This is standard in marketing mail and it works through your IP and user agent.
- Fonts and media. Font loading and embedded media can involve external fetches or a codec attack surface.
What actually happens when you double-click
The honest answer: it depends entirely on the application, which is why "opening a PDF" is not one risk level.
- Browser built-ins (Edge, Chrome) run in a sandbox with a heavily restricted PDF JavaScript implementation. Risk is comparatively low, and the sandbox is the reason.
- Full editors (Acrobat and similar) implement the widest feature set, including JavaScript and forms — with a correspondingly wider surface, mitigated by Protected Mode and sandboxing in current versions.
- Minimal native viewers that deliberately implement neither JavaScript nor action execution remove those classes of risk outright. What they cannot do is protect you from a parser vulnerability or from a file you then hand to another application.
None of these protect you from you: opening an embedded attachment, or clicking a link inside the document, is still opening an attachment or visiting a site.
Which protections actually matter
Ranked roughly by how much real-world risk they remove:
- No JavaScript execution. Eliminates the most abused PDF feature for delivery. A viewer that does not ship a JS engine cannot be tricked by one.
- Blocking automatic launch and submit actions. Even without JS, a document can specify an action on open. Nothing should happen without your click.
- No automatic network fetches. Blocks opening-tracking beacons and remote content the document tries to load without asking.
- Explicit confirmation before opening links or attachments. The user, not the document, decides.
- Sandboxing and memory-safe parsing. Mitigates parser bugs — the class of problem you cannot design away, only contain.
- Updatable. A viewer that never gets patched turns today's disclosed bug into a permanent hole.
Practical habits that help more than any setting
- Get documents from a known source. Most successful PDF-borne malware is a social-engineering problem first and a file-format problem second.
- Be suspicious of unexpected invoices, receipts and delivery notices — the three most common lures, precisely because people open them without thinking.
- Widen the extension.
invoice.pdf.exelooks likeinvoice.pdfwhen Windows hides known extensions. Turn that off. - Watch for a document that immediately asks you to enable content. A PDF prompting you to allow scripting or external content is a PDF asking you to turn off the protection.
- Scan before opening if the source is questionable; every mainstream AV engine handles PDFs.
- Keep the viewer updated. Unpatched parser bugs are the one risk no feature list can design away.
What FlyPDF does
- No PDF JavaScript engine — scripting is not implemented, so there is nothing to enable.
- No automatic launch or submit actions; no automatic network requests from a document.
- Links and embedded attachments require explicit confirmation before anything opens.
- Read-only by design: the viewer does not write to the document, which also removes the "modified original" problem covered in why read-only matters.
- No account, no telemetry, offline operation — opening a document produces no outbound traffic from the application itself.
- Private mode for working through sensitive material without leaving it in the recent-files list.
What it does not do, so you are not relying on the wrong thing
- It is not an antivirus. A malicious attachment you choose to open will still run. Confirmation prompts are a speed bump, not a security boundary.
- It does not verify signatures or tell you whether a document was tampered with before you received it.
- It does not detect phishing by content. A very convincing fake invoice still looks like an invoice.
- It does not sanitise files. If you need to strip active content before forwarding a document, use a dedicated cleaning tool or a PDF sanitiser in your mail gateway.
The one-line answer
Opening a PDF from a trusted source in an updated viewer that does not run JavaScript is low risk and normal. Opening an unexpected PDF in software that implements the full interactive feature set is a different proposition — and the single most valuable thing a viewer can do is simply not implement the parts that were never needed for reading.
<!-- fd:hublink --> Related guide: see PDF password protection for a detailed walkthrough.
FlyPDF does this locally. Read-only by design: OCR, search, annotations and text export all run on your machine, and the source file is never rewritten.
Try FlyPDF free for 30 days