Oracle WebCenter Content EUVD-2026-46991

Exploitation requires that a victim who is a legitimate user of the target Oracle WebCenter Content instance - 'a person other than the attacker' per the description - actively interact with attacker-supplied content (e.g. click a crafted link or open crafted content) while reachable over HTTP; this UI:R requirement is the primary limiting factor. The attacker needs no credentials or prior privileges on the target (PR:N) and no special product configuration is named in the description, so default deployments of versions 12.2.1.4.0 and 14.1.2.0.0 are exposed, but exploitation is bounded by network HTTP reachability to the Content Server and the presence of an interactable, typically authenticated, user session to abuse.

The 8.8 CVSS base score is driven by network reach, low attack complexity and no privilege requirement (AV:N/AC:L/PR:N), but the single most important mitigating metric is UI:R - a person other than the attacker must be induced to act, which converts this from an auto-exploitable server bug into a socially-engineered client-side attack and materially lowers real-world weaponizability. No EPSS score, SSVC decision, KEV entry, or POC was provided, so probabilistic exploitation signals are entirely absent and cannot be weighed. Scope is Unchanged (S:U), meaning impact is contained to the WebCenter Content instance rather than propagating to the host. Net assessment: a genuine high-severity issue warranting prompt patching for internet- or intranet-exposed instances, but not an emergency drop-everything item absent evidence of active exploitation; the requirement for victim interaction and the lack of any exploit signal suggest targeted rather than mass exploitation is the realistic threat.

An attacker crafts a malicious web page or email link and lures an authenticated WebCenter Content user (for example a content administrator) into visiting it while their session is active; the crafted request is processed by the Content Server in the victim's security context, granting the attacker control over the content repository. The low attack complexity means the request itself is trivial to construct, but success hinges entirely on the social-engineering step. No public proof-of-concept is currently identified.