CVE-2026-0761
CriticalSummary
Podatność CVE-2026-0761 dotyczy wstrzykiwania kodu w funkcji actionoutput_str_to_mapping w Foundation Agents MetaGPT, co pozwala zdalnym atakującym na wykonanie dowolnego kodu. Wykorzystanie tej podatności nie wymaga uwierzytelnienia.
Risk Assessment
Zdalne wykonanie kodu może prowadzić do przejęcia kontroli nad systemem, co stwarza poważne zagrożenie dla bezpieczeństwa organizacji. Atakujący może działać w kontekście konta usługi, co zwiększa ryzyko.
Recommendation
Zaleca się natychmiastowe zaktualizowanie oprogramowania Foundation Agents MetaGPT do najnowszej wersji, aby usunąć tę podatność. Należy również wdrożyć dodatkowe mechanizmy zabezpieczeń, aby ograniczyć dostęp do krytycznych funkcji.
Related vulnerabilities
- CVE-2026-115995Critical
A vulnerability was detected in Unitree Robotics Unitree Go2 up to version 1.1.11.4. The set_volume function in gpt_thread.py of the chat_go component mishandles the Value argument, leading to OS command injection. The attack can be launched remotely, and upgrading to version 1.1.11.5 resolves the issue.
- CVE-2026-114820Critical
Vulnerability in the JSSE component of Oracle Java SE, Oracle GraalVM for JDK, and Oracle GraalVM Enterprise Edition. It is easily exploitable by an unauthenticated attacker with network access, leading to unauthorized creation, deletion, or modification of critical data as well as unauthorized access to critical data. The vulnerability also applies to Java deployments that load and run untrusted code in a sandbox (e.g., Java Web Start applications or applets).
- CVE-2026-115933Critical
Postiz signs its session, invitation and password reset tokens with one shared secret and without any marker of what a token is for. A route that refreshes the credentials of a browser-extension channel signed data taken from the request itself and stored the result where the requesting user can read it back, allowing a user to obtain a validly signed token containing any claims they choose, accepted as a session, an organization invitation with a chosen role, a password reset or an enterprise provisioning token.
- CVE-2026-115557Critical
Python has added a denylist for known compromised certificates to the "ssl" module to ensure user safety even without certificate revocation configured. Known compromised certificates will automatically be rejected regardless of the SSLContext policy configuration.
- CVE-2026-115932Critical
In Postiz, a media path that is not a web address is treated as a file on the server's disk without verifying it lies inside the upload directory. A registered user can store a crafted media path with a post and cause the publishing worker to read an arbitrary file and send it as an attachment to a Discord server the user controls. The same unconfined reads are reachable through two further routes, including file existence checks and backend memory exhaustion; self-registration is open by default.
- CVE-2026-94075Critical
The WordPress plugin Newsletters (newsletters-lite) up to version 4.18.1 contains a deserialization of untrusted data vulnerability, allowing object injection. This could enable an attacker to execute unintended code.
- CVE-2026-114196Critical
The WordPress plugin Paytium up to version 5.0.5 is vulnerable to deserialization of untrusted data, allowing object injection. This may enable an attacker to perform unauthorized actions.
- CVE-2026-112866Critical
The WordPress plugin Product Filter by WBW (woo-product-filter) up to version 3.4.6 is vulnerable to blind SQL injection due to improper neutralization of special elements in SQL commands. An attacker could exploit this to manipulate database queries.
- CVE-2026-111402Critical
The WordPress plugin Newsletters (newsletters-lite) up to version 4.18.1 is vulnerable to SQL injection due to improper neutralization of special elements in SQL commands. This could allow an attacker to execute arbitrary SQL queries.
- CVE-2026-110956Critical
The asVereinbase WordPress plugin before version 1.1.8 does not require a valid password or a genuine second factor before authenticating a user, treating an empty stored value as a match. This allows an unauthenticated attacker to log in as any user, including an administrator.
Original NVD description (English source)
Foundation Agents MetaGPT actionoutput_str_to_mapping Code Injection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of Foundation Agents MetaGPT. Authentication is not required to exploit this vulnerability. The specific flaw exists within the actionoutput_str_to_mapping function. The issue results from the lack of proper validation of a user-supplied string before using it to execute Python code. An attacker can leverage this vulnerability to execute code in the context of the service account. Was ZDI-CAN-28124.
Vulnerability data from NVD (NIST) · CISA KEV · EPSS

