Media Library Assistant CVE-2026-97293

Requires an authenticated WordPress account with at least Contributor privileges on a site running Media Library Assistant version 3.41 or earlier; exploitation is remote over the network with no user interaction. The practical limiting factor is obtaining that Contributor account - realistic on sites with open registration, multi-author blogs, or membership/guest-post configurations, but not on locked-down single-admin installs. No non-default plugin configuration is stated as necessary; the injection targets the plugin's media/attachment query handling.

CVSS 3.1 base score is 8.5 (High), driven by network attack vector (AV:N), low complexity (AC:L), and a scope change (S:C) that pushes confidentiality to C:H alongside a minor availability impact (A:L). Critically, PR:L means this is NOT unauthenticated - an attacker needs a valid WordPress account at Contributor role or higher, which materially limits the exploit population to sites that permit self-registration or grant contributor accounts. No EPSS score, no CISA KEV listing, and no public POC are present in the supplied data, so there is no evidence of active or imminent mass exploitation - this reads as a real-but-not-emergency issue for affected sites. The SQLi is genuine and the C:H disclosure (full database read, including password hashes and secret keys) makes it a meaningful privilege-escalation stepping stone, but the contributor-authentication prerequisite keeps real-world risk below the raw 8.5 headline. No conflicting signals were found between the vector and the description; both agree on an authenticated network-exploitable SQL injection.

An attacker registers or obtains a Contributor-level WordPress account on a site running Media Library Assistant <= 3.41, then submits a crafted request to a plugin endpoint (e.g., media/attachment query parameters) containing malicious SQL. Because input is concatenated into a database query, the attacker exfiltrates sensitive data such as user password hashes and WordPress secret keys, enabling further account takeover. No public exploit code is identified at time of analysis, though Patchstack-reported plugin SQLi of this class is typically straightforward to weaponize.