| CVE |
Vendors |
Products |
Updated |
CVSS v3.1 |
| 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. |
| Rox, the software running BeWelcome, contains a PHP object injection vulnerability resulting from deserialization of untrusted data. User-controlled input is passed to PHP's unserialize(): the POST parameter `formkit_memory_recovery` in \\RoxPostHandler::getCallbackAction and the 'memory cookie' read by \\RoxModelBase::getMemoryCookie (bwRemember). (1) If present, `formkit_memory_recovery` is processed and passed to unserialize(), and (2) restore-from-memory functionality calls unserialize() on the bwRemember cookie value. Gadget chains present in Rox and bundled libraries enable exploitation of object injection to write arbitrary files or achieve remote code execution. Successful exploitation can lead to full site compromise. This vulnerability was remediated with commit c60bf04 (2025-06-16). |
| Deserialization of Untrusted Data vulnerability in strongholdthemes Dental Care CPT dentalcare-cpt allows Object Injection.This issue affects Dental Care CPT: from n/a through <= 20.2. |
| Deserialization of Untrusted Data vulnerability in designthemes Solar Energy solar allows Object Injection.This issue affects Solar Energy: from n/a through <= 3.5. |
| Deserialization of Untrusted Data vulnerability in Jthemes Prestige prestige allows Object Injection.This issue affects Prestige: from n/a through < 1.4.1. |
| Deserialization of Untrusted Data vulnerability in wpshuffle Subscribe to Download subscribe-to-download allows Object Injection.This issue affects Subscribe to Download: from n/a through <= 2.0.9. |
| The Keras.Model.load_model method, including when executed with the intended security mitigation safe_mode=True, is vulnerable to arbitrary local file loading and Server-Side Request Forgery (SSRF).
This vulnerability stems from the way the StringLookup layer is handled during model loading from a specially crafted .keras archive. The constructor for the StringLookup layer accepts a vocabulary argument that can specify a local file path or a remote file path.
* Arbitrary Local File Read: An attacker can create a malicious .keras file that embeds a local path in the StringLookup layer's configuration. When the model is loaded, Keras will attempt to read the content of the specified local file and incorporate it into the model state (e.g., retrievable via get_vocabulary()), allowing an attacker to read arbitrary local files on the hosting system.
* Server-Side Request Forgery (SSRF): Keras utilizes tf.io.gfile for file operations. Since tf.io.gfile supports remote filesystem handlers (such as GCS and HDFS) and HTTP/HTTPS protocols, the same mechanism can be leveraged to fetch content from arbitrary network endpoints on the server's behalf, resulting in an SSRF condition.
The security issue is that the feature allowing external path loading was not properly restricted by the safe_mode=True flag, which was intended to prevent such unintended data access. |
| A vulnerability classified as critical has been found in thu-pacman chitu 0.1.0. This affects the function torch.load of the file chitu/chitu/backend.py. The manipulation of the argument ckpt_path/quant_ckpt_dir leads to deserialization. An attack has to be approached locally. |
| A PHP object injection vulnerability exists in SugarCRM versions prior to 6.5.24, 6.7.13, 7.5.2.5, 7.6.2.2, and 7.7.1.0 due to improper validation of PHP serialized input in the SugarRestSerialize.php script. The vulnerable code fails to sanitize the rest_data parameter before passing it to the unserialize() function. This allows an unauthenticated attacker to submit crafted serialized data containing malicious object declarations, resulting in arbitrary code execution within the application context. Although SugarCRM released a prior fix in advisory sugarcrm-sa-2016-001, the patch was incomplete and failed to address some vectors. Exploitation evidence was observed by the Shadowserver Foundation on 2024-09-13 UTC. |
| The Meta Tag Manager plugin for WordPress is vulnerable to PHP Object Injection in all versions up to, and including, 3.0.2 via deserialization of untrusted input in the get_post_data function. This makes it possible for authenticated attackers, with contributor access or higher, to inject a PHP Object. No POP chain is present in the vulnerable plugin. If a POP chain is present via an additional plugin or theme installed on the target system, it could allow the attacker to delete arbitrary files, retrieve sensitive data, or execute code. |
| Deserialization of untrusted data can occur in versions 0.6 or newer of the skops python library, enabling a maliciously crafted model to run arbitrary code on an end user's system when loaded. |
| The sr_feuser_register extension through 12.4.8 for TYPO3 allows Remote Code Execution. |
| Deserialization of Untrusted Data vulnerability in BoldThemes Travelicious travelicious allows Object Injection.This issue affects Travelicious: from n/a through < 1.6.7. |
| A vulnerability was discovered in the Arctera InfoScale 7.0 through 8.0.2 where a .NET remoting endpoint can be exploited due to the insecure deserialization of potentially untrusted messages. The vulnerability is present in the Windows Plugin_Host service, which runs on all the servers where InfoScale is installed. The service is used only when applications are configured for Disaster Recovery (DR) using the DR wizard. Disabling the Plugin_Host service manually will eliminate the vulnerability. |
| Insecure deserialization in Hodoku v2.3.0 to v2.3.2 allows attackers to execute arbitrary code. |
| Cap Collectif is an online decision making platform that integrates several tools. Before commit 812f2a7d271b76deab1175bdaf2be0b8102dd198, the `DebateAlternateArgumentsResolver` deserializes a `Cursor`, allowing any classes and which can be controlled by unauthenticated user. Exploitation of this vulnerability can lead to Remote Code Execution. The vulnerability is fixed in commit 812f2a7d271b76deab1175bdaf2be0b8102dd198. |
| An issue was discovered in RDoc 6.3.3 through 6.6.2, as distributed in Ruby 3.x through 3.3.0. When parsing .rdoc_options (used for configuration in RDoc) as a YAML file, object injection and resultant remote code execution are possible because there are no restrictions on the classes that can be restored. (When loading the documentation cache, object injection and resultant remote code execution are also possible if there were a crafted cache.) The main fixed version is 6.6.3.1. For Ruby 3.0 users, a fixed version is rdoc 6.3.4.1. For Ruby 3.1 users, a fixed version is rdoc 6.4.1.1. For Ruby 3.2 users, a fixed version is rdoc 6.5.1.1. |
| CWE-502: Deserialization of Untrusted Data vulnerability exists that could allow code to be
remotely executed on the server when unsafely deserialized data is posted to the web server. |
| A deserialization vulnerability exists in BentoML's runner server in bentoml/bentoml versions <=1.3.4.post1. By setting specific parameters, an attacker can execute unauthorized arbitrary code on the server, causing severe harm. The vulnerability is triggered when the args-number parameter is greater than 1, leading to automatic deserialization and arbitrary code execution. |
| NVIDIA TensorRT-LLM for any platform contains a vulnerability in python executor where an attacker may cause a data validation issue by local access to the TRTLLM server. A successful exploit of this vulnerability may lead to code execution, information disclosure and data tampering. |
