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...

Holmes: An Evidence-Grounded LLM Agent for Auditable DDoS Investigation in Cloud Networks

Cloud environments face frequent DDoS threats due to centralized resources and broad attack surfaces. Modern cloud-native DDoS attacks further evolve rapidly and often blend multi-vector strategies, creating an operational dilemma: defenders need wire-speed monitoring while also requiring...

CPython security vulnerabilities

CPython is a Python interpreter implemented in C language by the Python Foundation. CPython has security vulnerabilities, which stem from the possibility of injecting additional commands through line breaks when user-controlled commands are passed...

Rethinking On-Device LLM Reasoning: Why Analogical Mapping Outperforms Abstract Thinking for IoT DDoS Detection

The rapid expansion of IoT deployments has intensified cybersecurity threats, notably Distributed Denial of Service DDoS attacks, characterized by increasingly sophisticated patterns. Leveraging Generative AI through On-Device Large Language Models ODLLMs provides a viable solution for real-time...

A Prompt-Based Framework for Loop Vulnerability Detection Using Local LLMs

Loop vulnerabilities are one major risky construct in software development. They can easily lead to infinite loops or executions, exhaust resources, or introduce logical errors that degrade performance and compromise security. The problem are often undetected by traditional static analyzers becau...

Linux Distros Unpatched Vulnerability : CVE-2026-23874

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - ImageMagick is free and open-source software used for editing and manipulating digital images. Versions prior to 7.1.2-13 have a stack overflow via infinite...

MiracleLinux 8 : firefox-91.2.0-4.el8.ML.1 (AXSA:2021-2461:28)

The remote MiracleLinux 8 host has a package installed that is affected by multiple vulnerabilities as referenced in the AXSA:2021-2461:28 advisory. Mozilla: Use-after-free in MessageTask CVE-2021-38496 Mozilla: Memory safety bugs fixed in Firefox 93, Firefox ESR 78.15, and Firefox ESR 91.2...

MiracleLinux 8 : container-tools:4.0 (AXSA:2023-7321:01)

The remote MiracleLinux 8 host has packages installed that are affected by a vulnerability as referenced in the AXSA:2023-7321:01 advisory. golang: net/http: insufficient sanitization of Host header CVE-2023-29406 Tenable has extracted the preceding description block directly from the MiracleLinu...

MiracleLinux 9 : mysql-8.0.41-2.el9_5.ML.1 (AXSA:2025-9701:03)

The remote MiracleLinux 9 host has packages installed that are affected by multiple vulnerabilities as referenced in the AXSA:2025-9701:03 advisory. openssl: SSLselectnextproto buffer overread CVE-2024-5535 krb5: GSS message token handling CVE-2024-37371 curl: libcurl: ASN.1 date parser overread...

PT-2026-3539

A vulnerability was detected in Bjskzy Zhiyou ERP up to 11.0. Impacted is the function initRCForm of the file RichClientService.class of the component com.artery.richclient.RichClientService. Performing a manipulation results in xml external entity reference. The attack is possible to be carried...

VulnResolver: A Hybrid Agent Framework for LLM-Based Automated Vulnerability Issue Resolution

As software systems grow in complexity, security vulnerabilities have become increasingly prevalent, posing serious risks and economic costs. Although automated detection tools such as fuzzers have advanced considerably, effective resolution still often depends on human expertise. Existing...

Constructing Multi-Label Hierarchical Classification Models for MITRE ATT&CK Text Tagging

MITRE ATT&CK is a cybersecurity knowledge base that organizes threat actor and cyber-attack information into a set of tactics describing the reasons and goals threat actors have for carrying out attacks, with each tactic having a set of techniques that describe the potential methods used in these...

MiracleLinux 9 : golang-1.20.12-4.el9_3 (AXSA:2024-7718:03)

The remote MiracleLinux 9 host has packages installed that are affected by a vulnerability as referenced in the AXSA:2024-7718:03 advisory. golang: net/http, x/net/http2: unlimited number of CONTINUATION frames causes DoS CVE-2023-45288 CVE-2023-45288 An attacker may cause an HTTP/2 endpoint to...

MiracleLinux 8 : dotnet3.1-3.1.422-1.el8.ML.1 (AXSA:2022-3776:10)

The remote MiracleLinux 8 host has packages installed that are affected by a vulnerability as referenced in the AXSA:2022-3776:10 advisory. dotnet: External Entity Injection during XML signature verification CVE-2022-34716 Tenable has extracted the preceding description block directly from the...

MiracleLinux 9 : buildah-1.31.3-1.el9 (AXSA:2023-6640:04)

The remote MiracleLinux 9 host has packages installed that are affected by multiple vulnerabilities as referenced in the AXSA:2023-6640:04 advisory. golang: html/template: improper handling of JavaScript whitespace CVE-2023-24540 net/http, golang.org/x/net/http2: avoid quadratic complexity in HPA...

MiracleLinux 9 : git-lfs-3.2.0-1.el9 (AXSA:2023-5844:01)

The remote MiracleLinux 9 host has a package installed that is affected by multiple vulnerabilities as referenced in the AXSA:2023-5844:01 advisory. golang: net/http: improper sanitization of Transfer-Encoding header CVE-2022-1705 golang: net/http/httputil: ReverseProxy should not forward...

CVE-2026-1179 Yonyou KSOA HTTP GET Parameter user_popedom.jsp sql injection

A vulnerability was detected in Yonyou KSOA 9.0. This affects an unknown part of the file /kmf/userpopedom.jsp of the component HTTP GET Parameter Handler. The manipulation of the argument folderid results in sql injection. The attack can be launched remotely. The exploit is now public and may be...

CVE-2026-23845

Mailpit is an email testing tool and API for developers. Versions prior to 1.28.3 are vulnerable to Server-Side Request Forgery SSRF via HTML Check CSS Download. The HTML Check feature /api/v1/message/ID/html-check is designed to analyze HTML emails for compatibility. During this process, the...

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...

CVE-2026-22850 Koko Analytics vulnerable to arbitrary SQL execution through unescaped analytics export/import and permissive admin SQL import

Koko Analytics is an open-source analytics plugin for WordPress. Versions prior to 2.1.3 are vulnerable to arbitrary SQL execution through unescaped analytics export/import and permissive admin SQL import. Unauthenticated visitors can submit arbitrary path pa and referrer r values to the public...