OpenImageIO EUVD-2026-30404

Exploitation requires the victim to process a malicious HEIF image file using an application or tool that utilizes OpenImageIO's HEIF decoder. The HEIF format support must be enabled in the OpenImageIO build (default configuration in standard builds). User interaction is mandatory (UI:P in CVSS vector) - the victim must actively open, import, convert, or otherwise process the crafted image file through an OpenImageIO-backed application. No authentication is required (PR:N) for exploitation once the file reaches the target system. Attack complexity is low (AC:L) meaning exploitation is reliable and repeatable once the malicious file is delivered. The vulnerability specifically triggers during HEIF subimage metadata parsing, so images that do not exercise multi-layer or multi-resolution HEIF features will not trigger the overflow. Exploitation is limited to local attack vector (AV:L), meaning the attacker must deliver the malicious file to the target system via social engineering, compromised file shares, supply chain insertion into asset repositories, or other file delivery mechanisms rather than exploiting over the network directly. Successful exploitation is confined to the security context of the OpenImageIO-using application process (scope unchanged: SC:N), not enabling privilege escalation to system level unless the application itself runs with elevated privileges.

CVSS 4.0 score of 8.5 (High) reflects significant potential impact (code execution with high confidentiality, integrity, and availability impact) but with notable exploitation barriers. CVSS vector AV:L indicates local attack vector requiring the malicious image file be delivered to the target system, while UI:P mandates user interaction (opening/processing the image). AC:L means exploitation is straightforward once these conditions are met. PR:N confirms no authentication is required. The confined scope (SC:N/SI:N/SA:N) limits impact to the vulnerable application context. Real-world risk depends heavily on deployment context: VFX studios and animation pipelines routinely process third-party image assets, making social engineering delivery plausible via email attachments, shared cloud folders, or compromised asset repositories. However, the local-only attack vector and user interaction requirement significantly reduce mass-exploitation risk compared to network-based vulnerabilities. No EPSS score provided in available data. Not listed in CISA KEV, indicating no confirmed widespread or targeted exploitation campaigns at this time. No public POC identified, though the detailed advisory description and availability of patch diffs may enable proof-of-concept development. Priority for organizations: High for VFX/animation studios and content pipelines processing untrusted HEIF images; Medium for general enterprise environments where OpenImageIO may be present but not processing external images. The combination of high impact and moderate exploitation difficulty warrants prompt patching in affected environments.

An attacker crafts a malicious HEIF image file with deliberately mismatched subimage metadata values designed to trigger the heap buffer overflow when parsed by OpenImageIO's HEIF decoder. The attacker delivers this weaponized image via email attachment to a VFX studio employee, disguised as a legitimate production asset or portfolio submission. When the employee opens the image in an application using OpenImageIO for rendering or conversion (such as a compositor, image viewer, or asset management tool), the HEIF decoder allocates insufficient heap memory based on one metadata field while writing data according to a larger conflicting metadata value, causing out-of-bounds writes that corrupt adjacent heap structures. The attacker leverages this memory corruption to overwrite function pointers or other control data, redirecting execution to attacker-controlled shellcode embedded in the image file, achieving arbitrary code execution with the privileges of the OpenImageIO-using application. No public POC exists at time of analysis, though the detailed technical description enables development by skilled attackers.