EUVD-2026-20928

A security flaw has been discovered in GL.iNet GL-RM1, GL-RM10, GL-RM10RC and GL-RM1PE 1.8.1. Affected by this issue is some unknown functionality of the component Factory Reset Handler. Performing a manipulation results in improper authentication. The attack can be initiated remotely. The...

ONNX: Malicious ONNX models can crash servers by exploiting unprotected object settings.

...

wasmtime 安全漏洞

Wasmtime is a lightweight WebAssembly runtime open source by the Bytecode Alliance. Versions prior to 24.0.7, 36.0.7, 42.0.2, and 43.0.1 of Wastime contain security vulnerabilities. These vulnerabilities arise when using the Val type promotion feature to increase values of component models. If bi...

CVE-2025-63238: Improper Neutralization of Input During Web Page Generation

A Reflected Cross-Site Scripting XSS affects LimeSurvey versions prior to 6.15.11+250909, due to the lack of validation of gid parameter in getInstance function in application/models/QuestionCreate.php. This allows an attacker to craft a malicious URL and compromise the logged in user...

CVE-2026-5815

A vulnerability was detected in D-Link DIR-645 1.01/1.02/1.03. Impacted is the function hedwigcgimain of the file /cgi-bin/hedwig.cgi. The manipulation results in stack-based buffer overflow. The attack can be launched remotely. The exploit is now public and may be used. This vulnerability only...

Allocation of Resources Without Limits or Throttling

Overview react-server-dom-parcel is a React Server Components bindings for DOM using Parcel. This is intended to be integrated into meta-frameworks. It is not intended to be imported directly. Affected versions of this package are vulnerable to Allocation of Resources Without Limits or Throttling...

Allocation of Resources Without Limits or Throttling

Overview react-server-dom-turbopack is a React Server Components bindings for DOM using Turbopack. This is intended to be integrated into meta-frameworks. It is not intended to be imported directly. Affected versions of this package are vulnerable to Allocation of Resources Without Limits or...

LobeHub: Unauthenticated authentication bypass on `webapi` routes via forgeable `X-lobe-chat-auth` header

Summary The webapi authentication layer trusts a client-controlled X-lobe-chat-auth header that is only XOR-obfuscated, not signed or otherwise authenticated. Because the XOR key is hardcoded in the repository, an attacker can forge arbitrary auth payloads and bypass authentication on protected...

GHSA-5MWJ-V5JW-5C97 LobeHub: Unauthenticated authentication bypass on `webapi` routes via forgeable `X-lobe-chat-auth` header

Summary The webapi authentication layer trusts a client-controlled X-lobe-chat-auth header that is only XOR-obfuscated, not signed or otherwise authenticated. Because the XOR key is hardcoded in the repository, an attacker can forge arbitrary auth payloads and bypass authentication on protected...

PT-2026-31562

Open AI models detect vulnerabilities by analyzing code, configs, and system behavior step-by-step—like spotting the FreeBSD zero-day CVE-2024-47467 in the Mythos showcase. Even a 3B open model nailed it, matching bigger ones. The chart shows rankings flip across tasks: OWASP easy stuff vs. gnarl...

EUVD-2026-19671

text-generation-webui is an open-source web interface for running Large Language Models. Prior to 4.3, he superbooga and superboogav2 RAG extensions fetch user-supplied URLs via requests.get with zero validation — no scheme check, no IP filtering, no hostname allowlist. An attacker can access clo...

CVE-2025-52908

An issue was discovered in the Wi-Fi driver in Samsung Mobile Processor and Wearable Processor Exynos 980, 850, 1280, 1330, 1380, 1480, 1580, W920, W930, and W1000. Incorrect Handling of the NL80211 vendor command leads to a buffer overflow via a certain ioctl message, issue 1 of 2...

Swiss-Bench 003: Evaluating LLM Reliability and Adversarial Security for Swiss Regulatory Contexts

The deployment of large language models LLMs in Swiss financial and regulatory contexts demands empirical evidence of both production reliability and adversarial security, dimensions not jointly operationalized in existing Swiss-focused evaluation frameworks. This paper introduces Swiss-Bench 003...

Guiding Symbolic Execution with Static Analysis and LLMs for Vulnerability Discovery

Symbolic execution detects vulnerabilities with precision, but applying it to large codebases requires harnesses that set up symbolic state, model dependencies, and specify assertions. Writing these harnesses has traditionally been a manual process requiring expert knowledge, which significantly...

CritBench: A Framework for Evaluating Cybersecurity Capabilities of Large Language Models in IEC 61850 Digital Substation Environments

The advancement of Large Language Models LLMs has raised concerns regarding their dual-use potential in cybersecurity. Existing evaluation frameworks overwhelmingly focus on Information Technology IT environments, failing to capture the constraints, and specialized protocols of Operational...

Stop Fixating on Prompts: Reasoning Hijacking and Constraint Tightening for Red-Teaming LLM Agents

With the widespread application of LLM-based agents across various domains, their complexity has introduced new security threats. Existing red-team methods mostly rely on modifying user prompts, which lack adaptability to new data and may impact the agent's performance. To address the challenge,...

Hackers or Hallucinators? A Comprehensive Analysis of LLM-Based Automated Penetration Testing

The rapid advancement of Large Language Models LLMs has created new opportunities for Automated Penetration Testing AutoPT, spawning numerous frameworks aimed at achieving end-to-end autonomous attacks. However, despite the proliferation of related studies, existing research generally lacks...

SkillSieve: A Hierarchical Triage Framework for Detecting Malicious AI Agent Skills

OpenClaw's ClawHub marketplace hosts over 13,000 community-contributed agent skills, and between 13% and 26% of them contain security vulnerabilities according to recent audits. Regex scanners miss obfuscated payloads; formal static analyzers cannot read the natural language instructions in...

Towards the Development of an LLM-Based Methodology for Automated Security Profiling in Compliance with Ukrainian Cybersecurity Regulations

In recent years, the pace of development of information technology in various areas has increased drastically, forcing cybersecurity specialists to constantly review existing processes in order to prevent unauthorized access to confidential information. Using Ukraine as a primary case study, this...

Stealthy and Adjustable Text-Guided Backdoor Attacks on Multimodal Pretrained Models

Multimodal pretrained models are vulnerable to backdoor attacks, yet most existing methods rely on visual or multimodal triggers, which are impractical since visually embedded triggers rarely occur in real-world data. To overcome this limitation, we propose a novel Text-Guided Backdoor TGB attack...