AI Engine EUVD-2026-114189

Exploitation requires an attacker to trick an authenticated WordPress user - in practice an administrator or other privileged operator with a valid session for the target site - into visiting attacker-controlled content, because the AI Engine plugin (through 3.8.8) fails to verify request authenticity for a state-changing action (CWE-352). The limiting factors are substantial: the victim must be actively logged in and must perform the interaction (UI:R), so there is no silent, fully remote unauthenticated path, and success depends on the victim holding sufficient privilege for the forced action to matter. The exact plugin endpoint/feature and the resulting action are not described in the available sources.

This is a CWE-352 Cross-Site Request Forgery flaw in the Meow Apps AI Engine WordPress plugin (through 3.8.8), reported by Patchstack and assigned CVSS 8.8 (CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H). The vector shows a network-reachable, low-complexity issue requiring no attacker privileges (PR:N) but mandatory victim interaction (UI:R) - the classic CSRF shape where the attacker holds no credentials and instead rides the session of a logged-in user who is lured into triggering the request. The high C/I/A ratings originate from Patchstack's scoring; the description does not state which plugin action is forced, so the actual impact is asserted by the source rather than demonstrated in the available text. vuln.today assigns a 'this_week' priority, indicating elevated but not emergency urgency. Key real-risk signals are absent: there is no CISA KEV listing, no EPSS probability was provided, and no exploitation status is indicated, so there is no evidence of active exploitation. The practical risk is therefore moderate and conditional - CSRF requires social-engineering a specific authenticated (typically administrator-level) victim into clicking, which bounds mass exploitation, but a popular AI plugin makes it an attractive opportunistic target. Treat the 8.8 as a severity ceiling that depends on landing user interaction, not as evidence of widespread automated attack.

An attacker crafts a malicious web page or email containing a forged request targeting the AI Engine plugin and lures a logged-in WordPress administrator into visiting it. Because the plugin does not validate request authenticity (missing/weak CSRF token), the victim's browser silently submits the state-changing request under their authenticated session. No public exploit code was identified at time of analysis, and the specific action the forged request performs is not described in the available sources.