Critical GPGME flaw lets a malicious OpenPGP key trigger a double free in apps that use it
Any application built on GPGME before 2.3.0 is exposed until it ships the patched library or a rebuilt copy, and the advisory offers no interim fix.
GnuPG's maintainers have disclosed a critical memory-corruption flaw, tracked as CVE-2026-129975, in GPGME, the library applications use to handle GnuPG operations. A malicious OpenPGP key carrying an unusually long user ID can trigger a double free, which the project says could be used to steal data from affected programs, though it has seen no attacks and cannot rule out remote code execution. Every application using GPGME before 2.3.0 is affected, whether it loads the library dynamically, as a Windows DLL, or statically, and the project names mail clients, XMPP clients, Samba and its own Gpg4win and GnuPG Desktop packages among the examples. The fixed release is GPGME 2.3.1, while Gpg4win 5.1.2 bundles the patched library and legacy 1.24.4 users have a separate release, and bundled or statically linked copies need their own rebuilt applications. The bug was reported by GitLab and disclosure was moved up from November 25 because of a separate Ruby2Shell issue, and the advisory does not describe an interim workaround for software that cannot yet be updated.
Why it matters
The flaw sits in a library that many mail, messaging and file-sharing programs rely on, so a single crafted key can reach software far beyond GnuPG itself. Users of those programs have no safe option until their vendors ship updates, since the project has not offered a workaround for software that cannot yet be patched. Because the project has seen no attacks but has not ruled out remote code execution, the practical concern is that exposure persists across every bundled or statically linked copy until each one is rebuilt.
Matters or hype?
Does this story matter, or is it hype? Vote before you see what everyone else said.