LLM Security and Safety: Insights from Homotopy-Inspired Prompt Obfuscation

In this study, we propose a homotopy-inspired prompt obfuscation framework to enhance understanding of security and safety vulnerabilities in Large Language Models LLMs. By systematically applying carefully engineered prompts, we demonstrate how latent model behaviors can be influenced in...

CVE-2026-1183: Improper Neutralization of Input During Web Page Generation

HTML injection vulnerability in multiple Botble products such as TransP, Athena, Martfury, and Homzen, consisting of an HTML injection due to a lack of proper validation of user input by sending a request to '/search' using the 'q' parameter...

Open5GS security vulnerabilities

Open5GS is an open-source implementation of 5G Core and EPC in C language, which serves as the core network for Lte/Nr networks. Open5GS has a security vulnerability, which stems from the use of a hardcoded JWT signing key...

MiracleLinux 9 : firefox-102.9.0-3.el9.ML.1 (AXSA:2023-5234:12)

The remote MiracleLinux 9 host has packages installed that are affected by multiple vulnerabilities as referenced in the AXSA:2023-5234:12 advisory. Mozilla: Incorrect code generation during JIT compilation CVE-2023-25751 Mozilla: Memory safety bugs fixed in Firefox 111 and Firefox ESR 102.9...

MiracleLinux 9 : tang-14-2.el9 (AXSA:2023-6769:02)

The remote MiracleLinux 9 host has a package installed that is affected by a vulnerability as referenced in the AXSA:2023-6769:02 advisory. tang: Race condition exists in the key generation and rotation functionality CVE-2023-1672 Tenable has extracted the preceding description block directly fro...

CVE-2026-23836

HotCRP is conference review software. A problem introduced in April 2024 in version 3.1 led to inadequately sanitized code generation for HotCRP formulas which allowed users to trigger the execution of arbitrary PHP code. The problem is patched in release version 3.2...

EUVD-2026-3322

Improper Neutralization of Input During Web Page Generation XSS or 'Cross-site Scripting' vulnerability in hexpm hexpm/hexpm 'Elixir.HexpmWeb.SharedAuthorizationView' modules allows Cross-Site Scripting XSS. This vulnerability is associated with program files...

Arbitrary Command Injection

@orval/mcp is vulnerable to Arbitrary Command Injection. The vulnerability is due to improper validation and escaping of the OpenAPI specification summary field during MCP server generation, which allows an attacker to break out of string literals and inject arbitrary code...

PT-2026-3443

Improper Neutralization of Input During Web Page Generation XSS or 'Cross-site Scripting' vulnerability in hexpm hexpm/hexpm 'Elixir.HexpmWeb.SharedAuthorizationView' modules allows Cross-Site Scripting XSS. This vulnerability is associated with program files lib/hexpm web/views/shared...

CVE-2025-67082

An SQL injection vulnerability in InvoicePlane through 1.6.3 has been identified in "maxQuantity" and "minQuantity" parameters when generating a report. An authenticated attacker can exploit this issue via error-based SQL injection, allowing for the extraction of arbitrary data from the database...

MiracleLinux 4 : ntp-4.2.6p5-5.0.1.AXS4 (AXSA:2015-327:04)

The remote MiracleLinux 4 host has packages installed that are affected by multiple vulnerabilities as referenced in the AXSA:2015-327:04 advisory. The Network Time Protocol NTP is used to synchronize a computer's time with another reference time source. This package includes ntpd a daemon which...

CVE-2026-0695: Improper Neutralization of Input During Web Page Generation

In ConnectWise PSA versions older than 2026.1, Time Entry notes stored in the Time Entry Audit Trail may be rendered without applying output encoding to certain content. Under specific conditions, this may allow stored script code to execute in the context of a user’s browser when the affected...

MiracleLinux 7 : httpd-2.4.6-89.0.1.el7.AXS7 (AXSA:2019-3965:02)

The remote MiracleLinux 7 host has packages installed that are affected by a vulnerability as referenced in the AXSA:2019-3965:02 advisory. httpd: Weak Digest auth nonce generation in modauthdigest CVE-2018-1312 Tenable has extracted the preceding description block directly from the MiracleLinux...

The vulnerability of the TAO software service for online modeling and optimization of AVEVA Process Optimization processes allows a hacker to execute arbitrary code and gain full control over the system.

The vulnerability of the TAO software service for online modeling and process optimization in AVEVA Process Optimization is related to improper code generation management. Exploiting this vulnerability allows a remote attacker to execute arbitrary code and gain full control over the system...

The vulnerability of software for online modeling and optimization of AVEVA Process Optimization processes, related to improper code generation management, allows attackers to execute arbitrary code, increase their privileges, and gain full control over the system.

The vulnerability of software for online simulation and optimization of AVEVA Process Optimization processes is related to improper code generation management. Exploiting this vulnerability allows attackers to execute arbitrary code, enhance their privileges, and gain full control over the system...

CVE-2026-23742: Improper Control of Generation of Code

Skipper is an HTTP router and reverse proxy for service composition. The default skipper configuration before 0.23.0 was -lua-sources=inline,file. The problem starts if untrusted users can create lua filters, because of -lua-sources=inline , for example through a Kubernetes Ingress resource. The...

Allocation of Resources Without Limits or Throttling

Overview Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling via the ACME TLS certificates' automatic generation. An attacker can exhaust system resources by opening multiple connections, sending minimal ClientHello messages with acme-tls/1, an...

Elon Musk’s Grok ‘Undressing’ Problem Isn’t Fixed

X has placed more restrictions on Grok’s ability to generate explicit AI images, but tests show that the updates have created a patchwork of limitations that fail to fully address the issue...

AppGuard Critiques AI Hyped Defenses; Expands its Insider Release for its Next-Generation Platform

McLean, Virginia, United States, 15th January 2026, CyberNewsWire...

CVE-2025-67082

InvoicePlane versions through 1.6.3 are vulnerable to an SQL injection vulnerability within the maxQuantity and minQuantity parameters used during report generation. The root cause is the insufficient sanitization of single quotes , which allows an authenticated attacker to perform error-based SQ...