GIMP EUVD-2026-104652

Exploitation requires a victim to open an attacker-crafted XCF (native GIMP) file in GIMP; the use-after-free is triggered specifically when a linked channel in the XCF loader's linked-channels list is destroyed via a selection property, leaving a dangling pointer that is subsequently dereferenced. Key limiting factors: the attack vector is LOCAL (AV:L) with no network-reachable component, mandatory user interaction is required (UI:R) - the file must be deliberately opened - and the impact is confined to the GIMP process running in the opening user's security context (S:U, so no privilege boundary is crossed). There are no special GIMP configuration prerequisites; any installation that can parse XCF files (the default) is affected, but a user must be socially engineered into opening the malicious file for exploitation to occur.

This is a genuine but conditionally-gated memory-safety defect rather than a drop-everything emergency. The provided CVSS 3.1 vector (AV:L/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H, base 7.8) is internally consistent with the description: the attack vector is LOCAL and requires user interaction (UI:R) because exploitation depends on a victim opening a crafted XCF file in GIMP - it is NOT a remotely-reachable network service. The use-after-free (CWE-416) yields high confidentiality, integrity, and availability impact, matching the described outcome of a crash and 'potentially' arbitrary code execution; note the vendor hedges on reliable RCE, so the most likely real-world outcome is a denial-of-service crash with code execution being harder to weaponize. No EPSS score, no CISA KEV listing, and no POC indicator were supplied, so there is no evidence of active or probable exploitation. The combination of local vector plus mandatory user interaction plus absence of any exploitation signal places this well below internet-facing or auto-exploitable flaws in priority: treat it as a standard-cadence patch for desktop/workstation fleets where GIMP is installed, prioritizing users who routinely open untrusted image files. No signals are in direct conflict; the main missing data is EPSS/KEV/POC status, whose absence (as noted below) is not itself evidence of safety.

An attacker distributes a maliciously crafted XCF file (e.g., via email attachment, a download, or a file-sharing site advertised as artwork or a project file). A victim opens it in GIMP, and the XCF loader's dangling pointer in the linked-channels list is dereferenced when the linked channel is destroyed via a selection property, triggering a use-after-free that crashes GIMP and could potentially be leveraged for code execution in the victim's user context. No public exploit identified at time of analysis, so weaponization into reliable RCE would require additional attacker development.