WebMaze: Defending Against LLM-Powered Web Scanners with Functionality-Coupled Deception

Large language models (LLMs) have enabled more effective vulnerability scanning of web applications. However, these capa

来源摘录 · arXiv cs.CR

GI-RTL-Fuzz: Grammar-Informed RTL Generation with LLMs for End-to-End Differential Simulator Fuzzing

Register-Transfer Level (RTL) simulators are fundamental to modern IC design workflows, yet like any complex software, t

来源摘录 · arXiv cs.CR

Manifest-Assisted Retrieval in Artificial Intelligence Search: Evidence Gaps and a Security Evaluation

The external sources retained in our non-systematic audit did not isolate the incremental security or utility effect of

来源摘录 · arXiv cs.CR

Is it my preference? Role Grounding Attack against Agent's Memory via Social Interaction

LLM-based agents now share experiences on social platforms and remember new preferences to enhance their performance. Ho

来源摘录 · arXiv cs.CR

Zero Trust-Enabled Adaptive Incident Response using Hierarchical Reinforcement Learning

Modern enterprises increasingly operate through complex information systems (IS) enabled by cloud computing, remote and

来源摘录 · arXiv cs.CR

Jailbreak Connectivity: Towards Diverse, Transferable, and Universal Multimodal Large Language Model Jailbreak

Multimodal large language models (MLLMs) are vulnerable to visual jailbreak attacks that bypass safety safeguards. Exist

来源摘录 · arXiv cs.CR

Understanding Post-Quantum Cryptography Adoption: Practices, Challenges, and Evolution in Open-Source Software

Post-quantum cryptography (PQC) standards establish algorithms for quantum-safe systems, but empirical evidence about ho

来源摘录 · arXiv cs.CR

Scan Before You Install: Detecting and Mitigating Local and Composite Vulnerabilities in Skills

An agent skill packages natural-language instructions, scripts, and reference files that an LLM-based agent loads for a

来源摘录 · arXiv cs.CR

Structured Security Explanations for Learning-Based Vulnerability Detection

Software vulnerability detection remains difficult because existing approaches rely on predefined analysis rules, learn

来源摘录 · arXiv cs.CR

Agentic Backdoors via Pretraining Poisoning

Large language models are pretrained on vast, uncurated corpora scraped from the public web. This creates a security ris

来源摘录 · arXiv cs.CR

The Envelope Layer: An Underestimated Surface of LLM Agent Defense

Prior work framed the envelope layer as the channel through which malicious content enters an agent, and established it

来源摘录 · arXiv cs.CR

Watermark Basins: Black-Box Detection of Distilled LLMs

Proprietary language models and their reasoning traces are vulnerable to API-based distillation, where an adversary trai

来源摘录 · arXiv cs.CR

When Skills Tell Agents Not to Ask: A Large-Scale Study of Oversight Directives in Agent Skills

Agent skills, reusable instruction packages centred on a SKILL.md file, extend AI agents on demand, and a single line in

来源摘录 · arXiv cs.CR

Blindfold: Keeping the Audit Coin Secret from LLM Agents

Audit-based certification bounds the violation rate of an LLM agent by auditing a random fraction $\varepsilon$ of actio

来源摘录 · arXiv cs.CR

IBM Langflow OSS 1.0.0 through 1.12.8 could allow a remote attacker to bypass security restrictions due to improper authentication.

CVE-2026-93680

IBM Langflow OSS 1.0.0 through 1.12.8 could allow a remote attacker to bypass security restrictions due to improper auth

来源摘录 · NVD / GitHub Advisory

An OS command injection vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows could allow arbitrary command execution under certain...

CVE-2026-90868

An OS command injection vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Wind

来源摘录 · NVD / GitHub Advisory

A path traversal vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows could allow a delegated Host Administrator to access locatio...

CVE-2026-90866

A path traversal vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows cou

来源摘录 · NVD / GitHub Advisory

An OS command injection vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows could allow arbitrary command execution with elevated...

CVE-2026-90865

An OS command injection vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Wind

来源摘录 · NVD / GitHub Advisory

A path traversal vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows could allow a delegated Host Administrator to access or modi...

CVE-2026-90864

A path traversal vulnerability in Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows cou

来源摘录 · NVD / GitHub Advisory

Missing authorization in the administrative interface of Progress Software WS_FTP Server versions prior to 2024.0.7 and 2025.0.4 on Windows could allow a delegated Host Administ...

CVE-2026-90863

Missing authorization in the administrative interface of Progress Software WS_FTP Server versions prior to 2024.0.7 and

来源摘录 · NVD / GitHub Advisory