| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| A remote code execution (RCE) vulnerability exists in the ms-swift project version 3.3.0 due to unsafe deserialization in tests/run.py using yaml.load() from the PyYAML library (versions = 5.3.1). If an attacker can control the content of the YAML configuration file passed to the --run_config parameter, arbitrary code can be executed during deserialization. This can lead to full system compromise. The vulnerability is triggered when a malicious YAML file is loaded, allowing the execution of arbitrary Python commands such as os.system(). It is recommended to upgrade PyYAML to version 5.4 or higher, and to use yaml.safe_load() to mitigate the issue. |
| The ParseAddressList function incorrectly handles comments (text within parentheses) within display names. Since this is a misalignment with conforming address parsers, it can result in different trust decisions being made by programs using different parsers. |
| A deserialization vulnerability exists in Volcengine's verl 3.0.0, specifically in the scripts/model_merger.py script when using the "fsdp" backend. The script calls torch.load() with weights_only=False on user-supplied .pt files, allowing attackers to execute arbitrary code if a maliciously crafted model file is loaded. An attacker can exploit this by convincing a victim to download and place a malicious model file in a local directory with a specific filename pattern. This vulnerability may lead to arbitrary code execution with the privileges of the user running the script. |
| The Debian zuluPolkit/CMakeLists.txt file for zuluCrypt through the zulucrypt_6.2.0-1 package has insecure PolicyKit allow_any/allow_inactive/allow_active settings that allow a local user to escalate their privileges to root. |
| OS Command Injection in iSTAR Ultra products web application allows an authenticated attacker to gain even more privileged access ('root' user) to the device firmware. |
| The LCD KVM over IP Switch CL5708IM has a Heap-based Buffer Overflow vulnerability in firmware versions prior to v2.2.215, allowing unauthenticated remote attackers to exploit this vulnerability to perform a denial-of-service attack. |
| The LCD KVM over IP Switch CL5708IM has a Heap-based Buffer Overflow vulnerability in firmware versions prior to v2.2.215, allowing unauthenticated remote attackers to exploit this vulnerability to perform a denial-of-service attack. |
| The LCD KVM over IP Switch CL5708IM has a Stack-based Buffer Overflow vulnerability in firmware versions prior to v2.2.215, allowing unauthenticated remote attackers to exploit this vulnerability to execute arbitrary code on the device. |
| In Akka through 2.10.6, akka-cluster-metrics uses Java serialization for cluster metrics. |
| iSTAR Ultra performs a firmware verification on boot, however the verification does not inspect certain portions of the firmware. These firmware parts may contain malicious code. Tested up to firmware 6.9.2, later firmwares are also possibly affected. |
| The modelscope/ms-swift library thru 2.6.1 is vulnerable to arbitrary code execution through deserialization of untrusted data within the `load_model_meta()` function of the `ModelFileSystemCache()` class. Attackers can execute arbitrary code and commands by crafting a malicious serialized `.mdl` payload, exploiting the use of `pickle.load()` on data from potentially untrusted sources. This vulnerability allows for remote code execution (RCE) by deceiving victims into loading a seemingly harmless checkpoint during a normal training process, thereby enabling attackers to execute arbitrary code on the targeted machine. Note that the payload file is a hidden file, making it difficult for the victim to detect tampering. More importantly, during the model training process, after the `.mdl` file is loaded and executes arbitrary code, the normal training process remains unaffected'meaning the user remains unaware of the arbitrary code execution. |
| An OS command injection vulnerability exists in Russound MBX-PRE-D67F firmware version 3.1.6, allowing unauthenticated attackers to execute arbitrary commands as root via crafted input to the hostname parameter in network configuration requests. This vulnerability stems from improper neutralization of special elements used in an OS command within the network configuration handler, enabling remote code execution with the highest privileges. |
| Paramount Macrium Reflect through 2025-06-26 allows attackers to execute arbitrary code with administrator privileges via a crafted .mrimgx or .mrbax backup file and a renamed executable placed in the same directory. When a user with administrative privileges opens the crafted backup file and proceeds to mount it, Reflect launches the renamed executable (e.g., explorer.exe), which is under attacker control. This occurs because of insufficient validation of companion files referenced during backup mounting. |
| A URL redirection in lbry-desktop v0.53.9 allows attackers to redirect victim users to attacker-controlled pages. |
| Paramount Macrium Reflect through 2025-06-26 allows local attackers to execute arbitrary code with administrator privileges via a crafted .mrimgx backup file and a malicious VSSSvr.dll located in the same directory. When a user with administrative privileges mounts a backup by opening the .mrimgx file, Reflect loads the attacker's VSSSvr.dll after the mount completes. This occurs because of untrusted DLL search path behavior in ReflectMonitor.exe. |
| DuraComm SPM-500 DP-10iN-100-MU
transmits sensitive data without encryption over a channel that could be intercepted by attackers. |
| Secure-upload is a data submission service that validates single-use tokens when accepting submissions to channels. The service only installed on a small number of environments.
Under specific circumstances, privileged users of secure-upload could have selected email templates not necessarily created for their enrollment when sending data upload requests.
Authenticated and privileged users of one enrollment could have abused an endpoint to redirect existing submission channels to a dataset they control.
An endpoint handling domain validation allowed unauthenticated users to enumerate existing enrollments.
Finally, other endpoints allowed enumerating if a resource with a known RID exists across enrollments.
The affected service has been patched with version 0.815.0 and automatically deployed to all Apollo-managed Foundry instances. |
| ModSecurity is an open source, cross platform web application firewall (WAF) engine for Apache, IIS and Nginx. In versions 2.9.8 to before 2.9.11, an empty XML tag can cause a segmentation fault. If SecParseXmlIntoArgs is set to On or OnlyArgs, and the request type is application/xml, and at least one XML tag is empty (eg <foo></foo>), then a segmentation fault occurs. This issue has been patched in version 2.9.11. A workaround involves setting SecParseXmlIntoArgs to Off. |
| A vulnerability in the password reset workflow of the Touch Lebanon Mobile App 2.20.2 allows an attacker to bypass the OTP reset password mechanism. By manipulating the reset process, an unauthorized user may be able to reset the password and gain access to the account without needing to provide a legitimate authentication factor, such as an OTP. This compromises account security and allows for potential unauthorized access to user data. |
| Himmelblau is an interoperability suite for Microsoft Azure Entra ID and Intune. Himmelblau versions 0.9.0 through 0.9.14 and 1.00-alpha are vulnerable to a privilege escalation issue when Entra ID group-based access restrictions are configured using group display names instead of object IDs. Starting in version 0.9.0, Himmelblau introduced support for specifying group names in the `pam_allow_groups` configuration option. However, Microsoft Entra ID permits the creation of multiple groups with the same `displayName` via the Microsoft Graph API—even by non-admin users, depending on tenant settings. As a result, a user could create a personal group with the same name as a legitimate access group (e.g., `"Allow-Linux-Login"`), add themselves to it, and be granted authentication or `sudo` rights by Himmelblau. Because affected Himmelblau versions compare group names by either `displayName` or by the immutable `objectId`, this allows bypassing access control mechanisms intended to restrict login to members of official, centrally-managed groups. This issue is fixed in Himmelblau version **0.9.15** and later. In these versions, group name matching in `pam_allow_groups` has been deprecated and removed, and only group `objectId`s (GUIDs) may be specified for secure group-based filtering. To mitigate the issue without upgrading, replace all entries in `pam_allow_groups` with the objectId of the target Entra ID group(s) and/or audit your tenant for groups with duplicate display names using the Microsoft Graph API. |
