GIMP EUVD-2026-94529

Exploitation requires a victim to open an attacker-supplied, very large image in GIMP and specifically invoke the 'Hot' color filter plug-in; the overflow only triggers when the image's width*height (and related) arithmetic overflows 32-bit integers, so only images with extremely large dimensions reach the vulnerable path. Limiting factors per the CVSS vector: local access (AV:L) with no remote network reachability, mandatory user interaction (UI:R) to load the file and run the filter, no privileges required (PR:N) but execution is confined to the running user's context, and high attack complexity (AC:H) because the attacker must construct dimensions that reliably overflow the allocation arithmetic. These constraints make mass or automated exploitation impractical; it is a targeted, social-engineering-dependent client-side attack.

This is a genuinely moderate-risk issue, not a high-priority emergency. The CVSS 3.1 base score of 6.3 (AV:L/AC:H/PR:N/UI:R/S:U/C:N/I:H/A:H) accurately reflects a constrained attack surface: the attacker needs local delivery of a file, the victim must actively load that file and invoke the Hot color filter, and the overflow condition only arises for 'very large images' where 32-bit width*height arithmetic overflows - hence the legitimate AC:H rating. No EPSS score was provided in the input, and KEV status is not indicated (treat as not listed). No proof-of-concept was referenced, so there is no public exploit identified at time of analysis. The impact is limited to integrity and availability (heap corruption leading to crash or potential code execution in the context of the desktop user running GIMP); confidentiality is rated N and there is no SSVC data supplied. The signals are internally consistent - a client-side, user-interaction-dependent memory-corruption bug in a desktop image editor - so there is no contradiction to flag between vector and description. The one caveat worth stating: a heap-based buffer overflow could in principle enable arbitrary code execution, which some analysts would score with C:H rather than C:N, so the confidentiality impact is arguably understated, but the provided vector's C:N is defensible if exploitation yields only corruption/crash.

An attacker distributes a maliciously crafted image file with extremely large dimensions (e.g. via email, a shared project, or a download) and social-engineers a GIMP user into opening it and applying the Hot color filter. The integer-overflowed allocation produces an undersized heap buffer while the filter's pixel path writes using the true image size, corrupting heap memory and potentially leading to code execution in the user's context. No public proof-of-concept exploit was identified in the provided data.