The vulnerability, described as an out-of-bounds write in the ICON decoder, was traced back to a specific commit in the ImageMagick repository. By comparing the git tags for the versions just before and after the patch was applied (7.1.2-24 and 7.1.2-25), I identified a commit with a message that directly referenced the security advisory GHSA-g22q-f7gc-5jhr. Analyzing this commit revealed a one-line change in the coders/icon.c file within the Read1XImage function. The loop condition was changed from iterating based on image->columns to image->rows. This incorrect loop boundary is the root cause of the heap buffer overflow. An attacker could craft an ICON file with a larger width than height, causing the loop to write data beyond the allocated memory for the image rows, leading to a denial of service or other potential impacts.