Sanitize and Scan Untrusted PDFs in .NET

Defanging PDFs from untrusted sources

Any app accepting uploaded PDFs (a portal, an applicant tracker, a claims system) can clean incoming files so embedded scripts and active content never reach a viewer:

using IronPdf;

var pdf = PdfDocument.FromFile("upload.pdf");
Cleaner.SanitizeWithSvg(pdf).SaveAs("upload-clean.pdf");

using IronPdf;

var pdf = PdfDocument.FromFile("upload.pdf");
Cleaner.SanitizeWithSvg(pdf).SaveAs("upload-clean.pdf");

Imports IronPdf

Dim pdf = PdfDocument.FromFile("upload.pdf")
Cleaner.SanitizeWithSvg(pdf).SaveAs("upload-clean.pdf")

Scanning before you process

ScanPdf checks a file against a default YARA ruleset (or a custom one) and reports whether anything dangerous was found, so a pipeline can quarantine or reject before storing or opening:

var pdf = PdfDocument.FromFile("upload.pdf");
CleanerScanResult result = Cleaner.ScanPdf(pdf);

if (result.IsDetected)
{
    // quarantine: result.Risks holds the detections
}

var pdf = PdfDocument.FromFile("upload.pdf");
CleanerScanResult result = Cleaner.ScanPdf(pdf);

if (result.IsDetected)
{
    // quarantine: result.Risks holds the detections
}

Imports System

Dim pdf = PdfDocument.FromFile("upload.pdf")
Dim result As CleanerScanResult = Cleaner.ScanPdf(pdf)

If result.IsDetected Then
    ' quarantine: result.Risks holds the detections
End If

It detects embedded JavaScript exploits, suspicious form actions, hidden content, known vulnerability patterns, and unauthorized embedded files. Note that ScanPdf only detects; cleaning is done by the SanitizeWith methods.

Hardening documents for cloud deployment

Stripping active content before a file enters a serverless or shared environment (Azure, AWS Lambda) reduces the attack surface there. Sanitize on the way in, not after the file already runs somewhere sensitive.

Stripping metadata before sharing

Because the document is rebuilt from an image, hidden author names, revision traces, and other metadata do not survive. Useful before publishing or sending a file beyond the organization.

Producing a clean baseline to sign or archive

Sanitizing before a digital signature or PDF/A archival gives you a script-free starting point. You rarely want to attest to, or preserve for decades, a document carrying active content.

Keeping text when you need it

When downstream steps need the words (search, indexing, accessibility, extraction), use SVG: it removes the active layer while keeping the PDF searchable, accepting possible layout drift.

What sanitization will not do

Sanitization is not redaction. It removes hidden and active threats, not visible information. Names, account numbers, and anything a human reads survive in full. To remove sensitive content, this is the wrong tool; use redaction.

And bitmap output is an image. It is unsearchable and inaccessible: a screen reader gets nothing from it, which conflicts with accessibility goals like PDF/UA. Reach for bitmap only when maximum security outweighs searchability, and use SVG when text must survive.

Treat it as a security step

Sanitization belongs to your security layer, not your formatting one. Run it on anything you did not generate yourself, scan first, keep the original, and pick bitmap or SVG by whether the text needs to live.