Oracle Agile PLM EUVD-2026-106114

Requires an attacker to hold a valid low-privileged authenticated account on an Oracle Agile PLM 9.3.6 instance reachable over HTTP, and requires human interaction from a different user (UI:R) - typically a victim viewing attacker-supplied content rendered by the Application Server - to trigger the scope-changing impact on additional products. Limiting factors that reduce real-world risk: exploitation cannot succeed against an unauthenticated outsider (PR:L authentication is mandatory), it depends on a second user interacting with the malicious content, and it is confined to the single affected version 9.3.6; impact is limited (partial read and partial write, no availability loss).

This is a moderate-severity, context-dependent issue rather than a top-tier emergency. The Oracle CVSS 3.1 base score is 5.4 (medium) with vector AV:N/AC:L/PR:L/UI:R/S:C/C:L/I:L/A:N: network attack vector and low complexity raise accessibility, but exploitation requires an existing low-privileged account (PR:L) AND human interaction from a different user (UI:R), which meaningfully constrains real-world weaponization. The low Confidentiality and Integrity impacts with no Availability impact, combined with the scope change (S:C), are the classic signature of a stored/reflected cross-site-scripting-style flaw in the application server where injected content runs in another user's authenticated context. No EPSS score, CISA KEV listing, POC, or SSVC/automatable data was supplied in the input, so there is no positive signal of active exploitation or automatable mass-exploitation - consistent with a medium CVSS that is NOT a drop-everything priority. Note a conflict: the 'Authentication Bypass' tag contradicts both the description and the CVSS (which requires PR:L authentication and UI:R), so that tag is likely mislabeled and should be verified rather than treated as fact. Overall: patch in the normal Oracle Critical Patch Update cycle; prioritize only if untrusted or weakly-trusted users can author content in the PLM instance.

A low-privileged but authenticated Agile PLM user injects a malicious payload into a data field that the application server later renders to another, higher-value user. When that victim views the content over HTTP, the payload executes in their authenticated session, allowing limited unauthorized read of some data and unauthorized update/insert/delete of a subset of data, with effects crossing into other products due to the scope change. No public exploit was identified at time of analysis, so an attacker would need to develop the payload themselves.