libheif, a library for decoding and encoding HEIF and AVIF images, has two bugs that allow a malicious AVIF/HEIC file to cause the decoder to copy uninitialized pixel planes during grid image decoding, exposing process heap memory as pixel values. An attacker who uploads such a file to any server‑side image processor that relies on libheif—such as WordPress, Sharp/libvips, or ImageMagick—can recover heap data, including function pointers sufficient to defeat ASLR or other secret information, from the publicly‑downloadable transcoded JPEG/PNG/WebP output. Local execution attacks are also possible, and the issue is fixed in libheif version 1.22.0.

The flaw exists in libheif versions earlier than 1.22.0. Any system that decodes HEIF or AVIF images using libheif—such as WordPress image handlers, Sharp/libvips, and ImageMagick—may be affected. Systems that accept uploaded HEIF/AVIF content and subsequently produce standard image outputs (JPEG, PNG, WebP) are at risk.

With a CVSS score of 7.5, the vulnerability carries high severity. The EPSS score of less than 1% indicates a low current exploitation probability, and it is not listed in the CISA KEV catalog. Exploitation requires an attacker to supply a crafted HEIF/AVIF file to a server‑side processor that performs a grid image decode and returns a transcoded image to the user, but local execution attacks are also possible. The result is a private‑data leak that could be leveraged to bypass ASLR and enable subsequent code execution techniques.