ImageMagick CVE-2026-116148

Exploitation requires that the surrounding application or workflow lets the attacker supply drawing options (drawing primitives/parameters) that are passed to ImageMagick's MVG encoder, and that the resulting MVG file is subsequently processed or rendered - the CVSS UI:P reflects this required downstream processing step. The encoder's failure to neutralize user-specified drawing options before writing them into the MVG output is the core precondition. Limiting factors that reduce realistic risk: the vector is local (AV:L), the attacker must already have a channel to inject drawing-option strings (so deployments that do not expose drawing options to untrusted input are not affected), the output must be consumed downstream for the injected commands to take effect, and only affected versions (6.x before 6.9.13-58, 7.x before 7.1.2-33) are vulnerable. The available sources do not describe the precise sink or what the injected MVG commands can ultimately achieve beyond command injection into the MVG file.

This is a medium-severity issue (CVSS 4.0 base 4.8) rather than a top-tier emergency, and the signals are internally consistent. The supplied CVSS 4.0 vector (AV:L/AC:L/AT:N/PR:N/UI:P/VC:L/VI:N/VA:N/SC:L/SI:N/SA:N) describes a local-vector flaw with low attack complexity, no privileges required against ImageMagick itself, but requiring user/pipeline interaction (UI:P) and yielding only Low confidentiality impact on both the vulnerable and a subsequent system - no integrity or availability impact in the vendor scoring. No EPSS score was provided, the CVE is not listed in CISA KEV, and the input shows only vendor/advisory references (GHSA and a VulnCheck advisory write-up), not a verified working exploit, so exploitation status is no public exploit identified at time of analysis. The vuln.today decision label is 'this_month', which matches a real-but-not-urgent priority: patch on a normal maintenance cadence rather than as an emergency. The key caveat is the trigger precondition - the CVE describes a second-order injection that only matters when an application lets an attacker control the drawing options passed to the MVG encoder and then processes the generated MVG output; that data-flow dependency, plus AV:L and UI:P, is what keeps real-world risk moderate despite ImageMagick's very large deployment footprint.

An application exposes ImageMagick drawing options (e.g., text, shapes, colors) to user input and uses them to produce an MVG file. An attacker embeds additional MVG drawing primitives in those options; because the MVG encoder writes the user-specified options into the output without neutralizing them, the injected commands become part of the generated MVG file and take effect when that file is subsequently processed or rendered. No public exploit identified at time of analysis; only vendor advisories are referenced.