Google Chrome CVE-2026-113106

Exploitation requires the victim to load attacker-controlled HTML/graphics content in a vulnerable Chrome desktop build prior to 156.0.8078.47 (UI:R - user must browse to the malicious page). No authentication or special privileges are needed (PR:N), and default configurations are affected since Skia rendering is core browser functionality. The primary LIMITING factor is that resulting code execution is confined to the sandboxed renderer process - full system compromise requires chaining a separate sandbox-escape vulnerability - and exploitation is gated on the user visiting malicious content rather than being fully automatable.

CVSS 3.1 base score is 8.8 (High) with vector AV:N/AC:L/PR:N/UI:R, indicating a network-reachable, low-complexity flaw needing no privileges but requiring user interaction (loading a crafted HTML page). However, Chromium's own security severity rating is only Medium, and critically the description states code execution occurs 'inside the sandbox' - meaning a successful exploit is confined to the renderer process and would need to be chained with a separate sandbox-escape bug to achieve full host compromise. This gap (CVSS 8.8 vs. vendor Medium) reflects that CVSS does not model Chrome's sandbox mitigation. No EPSS score, CISA KEV listing, or public POC is present in the provided data, so there is no evidence of active exploitation or available exploit code. Net assessment: patch promptly as part of normal Chrome update cadence, but this is not a drop-everything emergency - the sandbox containment and absence of exploitation signals place it below unsandboxed/actively-exploited Chrome zero-days. Skia (the graphics/rendering engine) is a historically fruitful exploitation surface, which is the main reason not to defer the update.

An attacker hosts or compromises a web page containing malicious graphics content that triggers the Skia memory-corruption bug when rendered. A victim using a vulnerable Chrome build (prior to 156.0.8078.47) visits the page, and the attacker gains arbitrary code execution within the sandboxed renderer process. No public exploit code or active exploitation is identified in the provided data, and the attacker would still need a separate sandbox escape to affect the underlying system.