CVE-2024-54142
CriticalSummary
Discourse AI to wtyczka do Discourse, która oferuje funkcje oparte na sztucznej inteligencji. W przypadku udostępniania rozmów z Discourse AI Bot w postach, HTML encje mogą wyciekać do aplikacji Discourse, gdy użytkownik odwiedza post z odnośnikiem do tej rozmowy.
Risk Assessment
Może to prowadzić do ujawnienia danych lub nieautoryzowanego dostępu do informacji w aplikacji Discourse, co stanowi zagrożenie dla bezpieczeństwa użytkowników.
Recommendation
Zaleca się aktualizację do najnowszej wersji, w której problem został rozwiązany. Użytkownicy, którzy nie mogą zaktualizować, powinni usunąć wszystkie grupy z ustawienia 'ai bot public sharing allowed groups'.
Related vulnerabilities
- CVE-2026-109694Critical
The Fluent CRM plugin (up to 3.2.5) contains an incorrect privilege assignment vulnerability that allows privilege escalation. An attacker could exploit this flaw to gain higher privileges within the WordPress system.
- CVE-2026-108497Critical
The Amelia Booking plugin (up to 2.4.10) contains an SQL Injection vulnerability due to improper neutralization of special elements in SQL commands. An attacker could execute arbitrary SQL queries, potentially leading to data leakage or modification.
- CVE-2026-50619Critical
A path traversal vulnerability has been reported in Qsirch, which could allow remote attackers to read unexpected files or system data. The vulnerability has been fixed in version 7.0.3.0 and later.
- CVE-2026-50612Critical
A Unix symbolic link (symlink) following vulnerability has been reported in QuMagie. If a remote attacker gains a user account, they can exploit the vulnerability to compromise the security of the system. The vulnerability has been fixed in version 2.9.2 and later.
- CVE-2026-47786Critical
An OS command injection vulnerability has been reported in HybridDesk Station, which could allow remote attackers to execute arbitrary commands. The vulnerability has been fixed in versions 4.2.23 and 4.3.4 and later.
- CVE-2026-34006Critical
A missing authentication for critical function vulnerability has been reported in sMedio DTCP MOVE, which could allow remote attackers to gain access to the system. The vulnerability has been fixed in version 1.4.5 and later.
- CVE-2026-19863Critical
In EthereumAuthenticator, an unauthenticated client may access certain protected resources if the request path contains the configured login path. This is caused by the authenticator using an unanchored substring match when identifying the login page, which may replace authentication or role requirements with ALLOWED.
- CVE-2026-16438Critical
Mitsubishi Electric products (GENESIS64, ICONICS Suite, Hyper Historian, AnalytiX, MobileHMI, MC Works64) versions 10.97.3 and prior contain an incorrect privilege assignment vulnerability. Installers grant default DCOM permissions to the administrator account, allowing an attacker to locally or remotely launch and operate DCOM objects that should require UAC elevation.
- CVE-2026-16436Critical
Mitsubishi Electric products (GENESIS, GENESIS64, ICONICS Suite, Hyper Historian, AnalytiX, MobileHMI, IoTWorX, MC Works64, GENESIS32, BizViz) contain a weak password recovery mechanism for forgotten passwords. The challenge response used to unlock the system is common to all users and fixed, allowing a remote attacker to unlock the system after obtaining it.
- CVE-2026-102469Critical
The Vert.x DB2 client (vertx-db2-client) deserializes User Defined Type (UDT) column data using a bare ObjectInputStream.readObject() call without an ObjectInputFilter or class allowlist. A malicious or man-in-the-middle DRDA server can send a response with a column tagged as UDT (FDOCA LID 0x50), causing deserialization of attacker-controlled bytes. TLS is disabled by default, making man-in-the-middle exploitation feasible.
Original NVD description (English source)
Discourse AI is a Discourse plugin which provides a number of AI features. When sharing Discourse AI Bot conversations into posts, if the conversation had HTML entities those could leak into the Discourse application when a user visited a post with a onebox to said conversation. This issue has been addressed in commit `92f122c`. Users are advised to update. Users unable to update may remove all groups from `ai bot public sharing allowed groups` site setting.
Vulnerability data from NVD (NIST) · CISA KEV · EPSS

