Inkscape EUVD-2026-16659

The CVSS:3.1 vector (AV:L/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:N) indicates this is a local attack vector requiring user interaction (opening a crafted SVG file) but no elevated privileges. The high confidentiality impact (C:H) signals that sensitive file contents can be disclosed. However, the local attack vector significantly constrains real-world exploitation scope compared to remote attacks; the attacker must either convince a user to open a malicious SVG or have local file access to place one. No EPSS score, CISA KEV status, or public exploit code has been identified in the provided intelligence, suggesting this is not an actively exploited vulnerability in the wild. The requirement for user interaction (UI:R) further reduces exploitation probability. From an SSVC perspective, this presents low to moderate exploitation risk outside targeted scenarios; organizations should prioritize patching based on Inkscape deployment scope and user exposure to untrusted SVG sources rather than treating it as an emergency. The patch availability (upstream fix in GitLab) and clear version boundary (pre-1.3 vs. 1.3+) make remediation straightforward.

An attacker sends a targeted user a malicious SVG file via email or embeds it on a website, crafted with xi:include tags pointing to sensitive local files such as SSH keys, configuration files, or credentials. When the user opens the file in Inkscape (meeting the UI:R requirement), the XInclude processor automatically resolves the file references and loads their contents into memory. The attacker can then extract sensitive data by observing error messages, timing behavior, or by exfiltrating rendered output. This attack is most effective in environments where users routinely work with SVG files from external or semi-trusted sources, such as design teams receiving assets from clients or contractors.