Intel

AIKIDO-2026-930803

pypdf is vulnerable to Denial of Service (DoS)

Denial of Service (DoS)GHSA-fc8x-2rww-xw9m Published Aug 10, 2026

48

Medium Risk

This Affects:

PYTHONpypdf
0.0.1 - 6.14.2
Fixed in 6.15.0
Are you affected? Scan for Free

TL;DR

pypdf scans PDF content for the next whitespace byte using read_until_whitespace while tokenizing documents. A crafted PDF that supplies a very long run of non-whitespace bytes drives an inefficient scan whose cost scales poorly with the input length. Processing such input consumes excessive CPU time and produces long runtimes. The fix reworks the scanning implementation so long non-whitespace inputs are handled efficiently.

Who does this affect?

You are affected if you are using a version that falls within the vulnerable range and you parse or extract content from PDF documents originating from untrusted or externally influenced sources, since tokenizing such input drives the read_until_whitespace scan.

Background info

pypdf is vulnerable to Denial of Service (DoS) in versions 0.0.1 - 6.14.2.

How to fix this

Upgrade the pypdf library to the patch version.