Anatomy of an Autonomous AI Agent Risk: How Qualys ETM Connects the Dots on OpenClaw

Executive Summary An unauthorized OpenClaw AI agent was detected disguised as a routine package on a Windows Server host. The situation escalated into a priority incident when Qualys ETM analyzed and correlated four distinct signals. While none of these signals alone warranted urgent action, the...

DeepGuard Secure Code Generation

Large Language Models LLMs for code generation can replicate insecure patterns from their training data. To mitigate this, a common strategy for security hardening is to fine-tune models using supervision derived from the final transformer layer. However, this design may suffer from a final-layer...

Wasmtime segfault or unused out-of-sandbox load with `f64x2.splat` operator on x86-64

On x86-64 platforms with SSE3 disabled Wasmtime's compilation of the f64x2.splat WebAssembly instruction with Cranelift may load 8 more bytes than is necessary. When signals-based-traps are disabled this can result in a uncaught segfault due to loading from unmapped guard pages. With guard pages...

GHSA-QQFJ-4VCM-26HV Wasmtime segfault or unused out-of-sandbox load with `f64x2.splat` operator on x86-64

On x86-64 platforms with SSE3 disabled Wasmtime's compilation of the f64x2.splat WebAssembly instruction with Cranelift may load 8 more bytes than is necessary. When signals-based-traps are disabled this can result in a uncaught segfault due to loading from unmapped guard pages. With guard pages...

CVE-2026-34971 Wasmtime miscompiled guest heap access enables sandbox escape on aarch64 Cranelift

Wasmtime is a runtime for WebAssembly. From 32.0.0 to before 36.0.7, 42.0.2, and 43.0.1, Wasmtime's Cranelift compilation backend contains a bug on aarch64 when performing a certain shape of heap accesses which means that the wrong address is accessed. When combined with explicit bounds checks a...

CVE-2026-39959 Tmds.DBus: malicious D-Bus peers can spoof signals, exhaust file descriptor resources, and cause denial of service

Tmds.DBus provides .NET libraries for working with D-Bus from .NET. Tmds.DBus and Tmds.DBus.Protocol are vulnerable to malicious D-Bus peers. A peer on the same bus can spoof signals by impersonating the owner of a well-known name, exhaust system resources or cause file descriptor spillover by...

MAL-2026-2520 Malicious code in @signals-notebook/utils (npm)

--- -= Per source details. Do not edit below this line.=- Source: amazon-inspector 6385e6085b941d851ce17c44dac94bb93521dad91d75b4d284a3dc8f9d367c2e The package @signals-notebook/utils was found to contain malicious code. Source: ossf-package-analysis...

PT-2026-31658

Tmds.DBus and Tmds.DBus.Protocol are vulnerable to malicious D-Bus peers. A peer on the same bus can spoof signals by impersonating the owner of a well-known name, exhaust system resources or cause file descriptor spillover by sending messages with an excessive number of Unix file descriptors, an...

CVE-2025-69515

An issue in JXL 9 Inch Car Android Double Din Player Android v12.0 allows attackers to force the infotainment system into accepting falsified GPS signals as legitimate, resulting in the device reporting an incorrect or static location...

JXL 9 Inch Car Android Double Din Player 安全漏洞

JXL 9 Inch Car Android Double Din Player is a vehicle infotainment system developed by JXL Corporation. Version 12.0 of the JXL 9 Inch Car Android Double Din Player contains a security vulnerability. This vulnerability arises from the ability for attackers to force the infotainment system to acce...

CVE-2026-5199

A writer role user in an attacker-controlled namespace could signal, delete, and reset workflows or activities in a victim namespace on the same cluster. Exploitation requires the attacker to know or guess specific victim workflow IDs and, for signal operations, signal names. This was due to a bu...

EUVD-2026-17995

A writer role user in an attacker-controlled namespace could signal, delete, and reset workflows or activities in a victim namespace on the same cluster. Exploitation requires the attacker to know or guess specific victim workflow IDs and, for signal operations, signal names. This was due to a bu...

On the Vulnerability of Deep Automatic Modulation Classifiers to Explainable Backdoor Threats

Deep learning DL has been widely studied for assisting applications of modern wireless communications. One of the applications is automatic modulation classification AMC. However, DL models are found to be vulnerable to adversarial machine learning AML threats. One of the most persistent and...

Identity security is the new pressure point for modern cyberattacks

Identity attacks no longer hinge on who a cyberattacker compromises, but on what that identity can access. As organizations manage growing numbers of human, non-human, and agentic identities, their access fabric multiplies across apps, resources, and environments, which increases both operational...

TLS Certificate and Domain Feature Analysis of Phishing Domains in the Danish .Dk Namespace

Phishing attacks remain a persistent cybersecurity threat, and the widespread adoption of TLS certificates has unintentionally enabled malicious websites to appear trustworthy to users. This study examines whether certificate metadata and domain characteristics can help distinguish phishing domai...

Secure agentic AI end-to-end

Next week, RSAC™ Conference celebrates its 35-year anniversary as a forum that brings the security community together to address new challenges and embrace opportunities in our quest to make the world a safer place for all. As we look towards that milestone, agentic AI is reshaping industries...

Everyday tools, extraordinary crimes: the ransomware exfiltration playbook

Data exfiltration activity increasingly leverages legitimate native utilities, commonly deployed third-party tools, and cloud service clients, reducing the effectiveness of static indicators of compromise IOCs and tool-based blocking strategies. The Exfiltration Framework systematically normalize...

Activation Surgery: Jailbreaking White-Box LLMs without Touching the Prompt

Most jailbreak techniques for Large Language Models LLMs primarily rely on prompt modifications, including paraphrasing, obfuscation, or conversational strategies. Meanwhile, abliteration techniques also known as targeted ablations of internal components have been used to study and explain LLM...

Understanding and Reducing AI Risk in Modern Applications

Identify real AI risk by connecting signals in context across the layers of AI applications...

LROO Rug Pull Detector: A Leakage-Resistant Framework Based on On-Chain and OSINT Signals

Smart contract-based ecosystems enable decentralized applications without trusted intermediaries, but their immutability and permissionless design also facilitate large-scale fraud. One of the most prevalent attacks is the rug pull, where project operators abruptly withdraw liquidity after...