CVE-2023-21639

Memory corruption in Audio while processing svamodelserializer using memory size passed by HIDL client...

CVE-2023-5245

FileUtil.extract enumerates all zip file entries and extracts each file without validating whether file paths in the archive are outside the intended directory. When creating an instance of TensorflowModel using the savedmodel format and an exported tensorflow model, the apply function invokes th...

CVE-2023-23080

Certain Tenda products are vulnerable to command injection. This affects Tenda CP7 Tenda CP7=V11.10.00.2211041403 and Tenda CP3 v.10 Tenda CP3 v.10=V202209060242025 and Tenda IT7-PCS Tenda IT7-PCS=V2209020914 and Tenda IT7-LCS Tenda IT7-LCS=V2209020914 and Tenda IT7-PRS Tenda IT7-PRS=V2209020908...

CVE-2023-42279

Dreamer CMS v4.1.3 was discovered to contain a SQL injection vulnerability via the model-form-management-field form...

CVE-2022-25505

Taocms v3.0.2 was discovered to contain a SQL injection vulnerability via the id parameter in \include\Model\Category.php...

CVE-2022-37709

Tesla Model 3 V11.02022.4.5.1 6b701552d7a6 Tesla mobile app v4.23 is vulnerable to Authentication Bypass by spoofing. Tesla Model 3's Phone Key authentication is vulnerable to Man-in-the-middle attacks in the BLE channel. It allows attackers to open a door and drive the car away by leveraging...

CVE-2022-40268

Improper Restriction of Rendered UI Layers or Frames vulnerability in Mitsubishi Electric Corporation GOT2000 Series GT27 model versions 01.14.000 to 01.47.000, Mitsubishi Electric Corporation GOT2000 Series GT25 model versions 01.14.000 to 01.47.000 and Mitsubishi Electric Corporation GT...

A Critical Evaluation of Defenses against Prompt Injection Attacks

Large Language Models LLMs are vulnerable to prompt injection attacks, and several defenses have recently been proposed, often claiming to mitigate these attacks successfully. However, we argue that existing studies lack a principled approach to evaluating these defenses. In this paper, we argue...

One Model Transfer to All: on Robust Jailbreak Prompts Generation against LLMs

Safety alignment in large language models LLMs is increasingly compromised by jailbreak attacks, which can manipulate these models to generate harmful or unintended content. Investigating these attacks is crucial for uncovering model vulnerabilities. However, many existing jailbreak strategies fa...

Gaming Tool Preferences in Agentic LLMs

Large language models LLMs can now access a wide range of external tools, thanks to the Model Context Protocol MCP. This greatly expands their abilities as various agents. However, LLMs rely entirely on the text descriptions of tools to decide which ones to use--a process that is surprisingly...

The vulnerability of the FUN_00459fdc function in the Totolink A3002R router’s microprogramming software allows a intruder to execute arbitrary commands.

The vulnerability of FUN00459fdc in the Totolink A3002R router’s microprogramming software relates to the failure to take measures to neutralize special elements used in the operating system commands. Exploiting this vulnerability allows a malicious actor to execute arbitrary commands remotely...

The vulnerability of the D-Link DI-8100 router’s microprogramming-based web service allows a hacker to circumvent existing security restrictions.

The vulnerability of the D-Link DI-8100 router microprogramming software web service is related to deficiencies in authentication procedures. Exploiting this vulnerability could allow a malicious actor to circumvent existing security restrictions remotely...

Chain-Of-Lure: a Synthetic Narrative-Driven Approach to Compromise Large Language Models

In the era of rapid generative AI development, interactions between humans and large language models face significant misusing risks. Previous research has primarily focused on black-box scenarios using human-guided prompts and white-box scenarios leveraging gradient-based LLM generation methods,...

ACSE-Eval: Can LLMs Threat Model Real-World Cloud Infrastructure?

While Large Language Models have shown promise in cybersecurity applications, their effectiveness in identifying security threats within cloud deployments remains unexplored. This paper introduces AWS Cloud Security Engineering Eval, a novel dataset for evaluating LLMs cloud security threat...

An Attack to Break Permutation-Based Private Third-Party Inference Schemes for LLMs

Recent advances in Large Language Models LLMs have led to the widespread adoption of third-party inference services, raising critical privacy concerns. Existing methods of performing private third-party inference, such as Secure Multiparty Computation SMPC, often rely on cryptographic methods...

The vulnerability of the sub_497DE4 function in D-Link DI-7003GV2 router microprogramming software allows a perpetrator to compromise the confidentiality, integrity, and accessibility of protected information.

The vulnerability of the sub497DE4 function in D-Link DI-7003GV2 router microprogramming software is related to deficiencies in authentication procedures. Exploiting this vulnerability could allow an attacker to compromise the confidentiality, integrity, and accessibility of the protected...

CVE-2022-21707

wasmCloud Host Runtime is a server process that securely hosts and provides dispatch for web assembly WASM actors and capability providers. In versions prior to 0.52.2 actors can bypass capability authorization. Actors are normally required to declare their capabilities for inbound invocations, b...

CVE-2022-40817

Zammad 5.2.1 has a fine-grained permission model that allows to configure read-only access to tickets. However, agents were still wrongly able to perform some operations on such tickets, like adding and removing links, tags. and related answers. This issue has been fixed in 5.2.2...

CVE-2022-44200

Netgear R7000P V1.3.0.8, V1.3.1.64 is vulnerable to Buffer Overflow via parameters: stamodedns1pri and stamodedns1sec...

CVE-2022-39893

Sensitive information exposure vulnerability in FmmBaseModel in Galaxy Buds Pro Manage prior to version 4.1.22092751 allows local attackers with log access permission to get device identifier data through device log...