
PDF red flags
Scan a PDF's raw bytes for tokens that indicate active content, JavaScript, auto-open actions, embedded files, incremental updates, as a quick triage signal, not proof of malice.
PDF red flags counts eight raw-byte tokens tied to active content, from /URI and /Launch through /JavaScript, /JS, /OpenAction, /AA, /EmbeddedFile and /RichMedia, tallying %%EOF markers for post-save edits. Five of those eight count as high risk on their own. Because this counts bytes rather than parsing real PDF objects, a token inside an unrelated literal string can trigger a false flag.
Questions
What tokens does it scan for?
/JavaScript, /JS, /OpenAction, /AA, /Launch, /EmbeddedFile, /URI and /RichMedia, counted across the raw bytes of the file.
Which of those count as high risk?
/JavaScript, /JS, /OpenAction, /Launch and /RichMedia; the rest only raise the verdict to worth a closer look.
What does the incremental-update note mean?
More than one %%EOF marker means the file has been edited and re-saved at least once since it was first written.
Is a clean verdict a guarantee of safety?
No, this is a byte-level token scan, not a real PDF object parser, so it can miss content and can also flag a token that sits inside a harmless literal string.
What does it check first?
That the file starts with the %PDF- header; anything else is refused as not a PDF.