ClawHub Security Signals: When VirusTotal, Static Analysis, and SkillSpector Disagree

Agent skills extend AI agents with reusable instructions, tools, scripts, references, and workflows, establishing a security boundary distinct from both model safety and traditional package-malware detection. ClawHub Security Signals is a sanitized dataset of 67,453 latest public OpenClaw skill...

A New Framework for Cybersecurity Refusals in AI Agents

Agentic scaffolds have dramatically improved LLM performance on complex, long-horizon tasks, yielding both broad benefits and amplified risks in domains like cybersecurity. Existing benchmarks for AI agents in cybersecurity focus mainly on measuring proficiency--how effectively agents can complet...

New Russia-Linked GREYVIBE Targets Ukraine with AI-Powered Cyberattacks

A previously undocumented threat actor dubbed GREYVIBE has been attributed to ongoing and persistent attacks targeting Ukraine and Ukraine-related entities since at least August 2025. GREYVIBE, per WithSecure, is assessed to be a Russian-speaking group operating broadly in the Russian time zone,...

BadBone: Backdoor Attacks against Backbone Models in Visual Prompt Learning

Prompt learning is a new machine learning paradigm that has attracted ample attention due to its simplicity and proven efficacy. Despite its growing adoption, the security vulnerabilities associated with this paradigm remain underexplored. In this work, we take the first step to propose BadBone, ...

CVE-2026-38702

A command injection vulnerability exists in the Admin Access feature of InHand Networks IR302 firmware V3.5.108, IR305 firmware V1.0.118, IR315 firmware V1.0.118, IR615 firmware V1.0.118, and earlier versions. Attackers can exploit this vulnerability to obtain ROOT privileges on remote target...

How Reliable Are AI Attackers against a Fixed Vulnerable Target? A 400-Run Empirical Study of LLM Penetration Testing Consistency

Large language models LLMs can autonomously conduct multi-stage cyber attacks, but the consistency of their offensive behavior under repeated trials remains unstudied. This work presents the first large-scale empirical measurement of LLM attack consistency: 400 autonomous penetration testing runs...

CVE-2026-38703

A command injection vulnerability exists in the ZeroTier VPN feature of InHand Networks IR302 firmware V3.5.108, IR305 firmware V1.0.118, IR315 firmware V1.0.118, IR615 firmware V1.0.118, and earlier versions. Attackers can exploit this vulnerability to obtain ROOT privileges on remote target...

Persona Attack: Incremental Memory Injection Jailbreak Attack against Large Language Models

As Large Language Models evolve for user convenience, vulnerability to jailbreak attacks continues to be reported despite ongoing efforts in safety training. Traditional jailbreak techniques typically focus on a single prompt injection, neglecting the models' ability to remember the flow of...

EUVD-2026-32278

In Slican telephone exchanges it is possible to manage the control panel remotely. An unauthenticated attacker can connect to the modem via a telephone with a specific caller ID. This allows them to bypass admin authentication and gain full access to the service protocol and configuration panel...

CVE-2026-35087

Slican telephone exchanges allow administrative protocol authentication bypass. An attacker can bypass the need to enter login credentials by executing the appropriate command. This issue was fixed in versions below: - NCP: version 1.24.0250 - IPx series: version 6.61.0040 - CCT-1668: version...

CVE-2026-35090

CVE-2026-35090 describes an authentication bypass in Slican telephone exchanges, allowing an unauthenticated attacker to remotely manage the control panel by dialing a specific caller ID. The issue enables bypass of admin authentication and full access to the service protocol and configuration pa...

CVE-2026-35087 Authentication Bypass in Slican telephone exchanges

Slican telephone exchanges allow administrative protocol authentication bypass. An attacker can bypass the need to enter login credentials by executing the appropriate command. This issue was fixed in versions below: - NCP: version 1.24.0250 - IPx series: version 6.61.0040 - CCT-1668: version...

CVE-2026-35087 Authentication Bypass in Slican telephone exchanges

Slican telephone exchanges allow administrative protocol authentication bypass. An attacker can bypass the need to enter login credentials by executing the appropriate command. This issue was fixed in versions below: - NCP: version 1.24.0250 - IPx series: version 6.61.0040 - CCT-1668: version...

Introducing EvidenceForge: Synthetic security logs that don’t look (as) fake

Security teams need high-quality, labeled datasets to train threat hunters and incident responders, validate detection logic, and develop robust analytic models. EvidenceForge helps teams overcome the limitations of anonymized or stale public datasets, while avoiding the cost and complexity of...

PT-2026-43700

In Slican telephone exchanges secure key is generated in a predictable manner using properties of the telephone exchange which can be obtained without authentication. An unauthenticated attacker can deduce the secure key and obtain admin credentials. This issue was fixed in versions below: - IPx...

S3C2 Summit 2025-07: Government Secure Supply Chain Summit

Software supply chains, while providing immense economic and software development value, are only as strong as their weakest link. Over the past several years, there has been an exponential increase in cyberattacks specifically targeting vulnerable links in critical software supply chains. The...

PT-2026-43701

In Slican telephone exchanges it is possible to manage the control panel remotely. An unauthenticated attacker can connect to the modem via a telephone with a specific caller ID. This allows them to bypass admin authentication and gain full access to the service protocol and configuration panel...

SAMD: A Tool for Identifying False Data Injection Scenarios in AI/ML-Enabled Medical Devices

The growing integration of artificial intelligence AI and machine learning ML in medical systems requires effective measures to address emerging security risks. One such risk is that of adversaries introducing false data through vulnerable system components during inference, causing misdiagnosis...

Relevance As a Vulnerability: How Web Retrieval Degrades Safety Alignment in LLM Agents

AI agents augment large language models with external tools such as web retrieval, enabling grounded and up-to-date responses. However, incorporating external content into the generation pipeline can weaken the safety alignment mechanisms that govern model outputs. Prior work shows that enabling...

MIRAGE: Context-Aware Prompt Injection against Mobile GUI Agents Via User-Generated Content

Mobile graphical user interface GUI agents driven by vision-language models VLMs perceive the screen as rendered pixels and choose actions from what they see, so they cannot reliably separate trusted interface elements from user-generated content. We present MIRAGE Mobile Injection of Realistic...