Google Chrome CVE-2026-106217

Exploitation requires the attacker to have ALREADY compromised the Chrome renderer process (a sandboxed-code-execution precondition, typically achieved via a separate chained exploit) - this is the single most important limiting factor and is not represented in the CVSS PR:N metric. Given that foothold, the attacker must get the victim to load a crafted HTML page (UI:R), and the flaw lies specifically in the Google Lens feature's missing authorization check (CWE-862), which permits reading cross-origin data. Limiting factors: no standalone remote exploitation (renderer must be compromised first), required user interaction to load the page, affected only on Chrome versions below 155.0.8059.39, and impact confined to low-sensitivity cross-origin information disclosure (no integrity or availability effect).

This is a genuinely low-priority, lower-real-risk issue despite its 'Medium' Chromium label. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:R/S:U/C:L/I:N/A:N = 4.3) scores confidentiality-only, low impact with no integrity or availability effect. Critically, the CVSS vector and the description conflict on prerequisites: the vector states PR:N (no privileges) yet the description explicitly requires 'a remote attacker who had compromised the renderer process' - that is a substantial precondition that the CVSS PR:N metric does not capture, because Chromium conventionally scores renderer-compromise chains as PR:N. In practice this bug is a secondary, post-exploitation primitive usable only after an attacker already controls a sandboxed renderer, not a standalone remote vulnerability. All exploitation signals point the same low-urgency direction: SSVC decision is Exploitation=none, Automatable=no, Technical Impact=partial; there is no CISA KEV listing and no public exploit identified at time of analysis. No EPSS score was provided in the input. The signals are consistent (low CVSS, benign SSVC, no KEV, no POC), so there is no conflict suggesting hidden urgency - the one caveat is the PR:N-versus-'compromised renderer' discrepancy noted above, which makes real-world exploitability narrower than the network vector implies.

An attacker who has already achieved code execution inside a victim's sandboxed Chrome renderer process (for example via a separate renderer exploit) serves or navigates the victim to a crafted HTML page that abuses the missing authorization check in Google Lens to read data belonging to another origin. The result is disclosure of cross-origin information the renderer should not be able to reach. No public exploit identified at time of analysis, and the UI:R metric means the victim must interact with (load) the malicious page.