AJAR: Adaptive Jailbreak Architecture for Red-Teaming

As Large Language Models LLMs evolve from static chatbots into autonomous agents capable of tool execution, the landscape of AI safety is shifting from content moderation to action security. However, existing red-teaming frameworks remain bifurcated: they either focus on rigid, script-based text...

PT-2026-3106

Name of the Vulnerable Software and Affected Versions SparkyFitness version 0.15.8.2 Description SparkyFitness is susceptible to Cross-Site Scripting XSS attacks. The issue stems from improper handling of user input and output from Large Language Models LLMs. This allows for the injection of...

SparkyFitness security vulnerability

SparkyFitness is a fitness and health management platform developed by CodeWithCJ. Version SparkyFitness v0.15.8.2 contains a security vulnerability, which stems from improper handling of user input and LLM outputs, potentially leading to cross-site scripting attacks...

EUVD-2026-1865

vLLM is vulnerable to DoS in Idefics3 vision models via image payload with ambiguous dimensions...

Integrating APK Image and Text Data for Enhanced Threat Detection: A Multimodal Deep Learning Approach to Android Malware

As zero-day Android malware attacks grow more sophisticated, recent research highlights the effectiveness of using image-based representations of malware bytecode to detect previously unseen threats. However, existing studies often overlook how image type and resolution affect detection and ignor...

opencode 安全漏洞

opencode is an AI programming intelligence open-sourced by Anomaly. A security vulnerability exists in versions prior to opencode 1.1.10, which stems from the Markdown renderer not cleaning up the LLM response, and could lead to the execution of JavaScript via HTML injection...

PYSEC-2026-143

vLLM is an inference and serving engine for large language models LLMs. In versions from 0.6.4 to before 0.12.0, users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimensi...

PYSEC-2026-143

vLLM is an inference and serving engine for large language models LLMs. In versions from 0.6.4 to before 0.12.0, users can crash the vLLM engine serving multimodal models that use the Idefics3 vision model implementation by sending a specially crafted 1x1 pixel image. This causes a tensor dimensi...

PT-2026-2260

Name of the Vulnerable Software and Affected Versions vLLM versions 0.6.4 through 0.11.9 Description vLLM is an inference and serving engine for large language models LLMs. Users can cause the vLLM engine to crash when serving multimodal models that utilize the Idefics3 vision model implementatio...

CVE-2025-62327

In HCL DevOps Deploy 8.1.2.0 through 8.1.2.3, a user with LLM configuration privileges may be able to recover a credential previously saved for performing authenticated LLM Queries...

CVE-2023-4899

SQL Injection in GitHub repository mintplex-labs/anything-llm prior to 0.0.1...

HogVul: Black-Box Adversarial Code Generation Framework against LM-Based Vulnerability Detectors

Recent advances in software vulnerability detection have been driven by Language Model LM-based approaches. However, these models remain vulnerable to adversarial attacks that exploit lexical and syntax perturbations, allowing critical flaws to evade detection. Existing black-box attacks on...

CVE-2026-21869

llama.cpp is an inference of several LLM models in C/C++. In commits 55d4206c8 and prior, the ndiscard parameter is parsed directly from JSON input in the llama.cpp server's completion endpoints without validation to ensure it's non-negative. When a negative value is supplied and the context fill...

Knowledge-To-Data: LLM-Driven Synthesis of Structured Network Traffic for Testbed-Free IDS Evaluation

Realistic, large-scale, and well-labeled cybersecurity datasets are essential for training and evaluating Intrusion Detection Systems IDS. However, they remain difficult to obtain due to privacy constraints, data sensitivity, and the cost of building controlled collection environments such as...

CVE-2025-62327

In HCL DevOps Deploy 8.1.2.0 through 8.1.2.3, a user with LLM configuration privileges may be able to recover a credential previously saved for performing authenticated LLM Queries...

CVE-2025-62327

In HCL DevOps Deploy 8.1.2.0 through 8.1.2.3, a user with LLM configuration privileges may be able to recover a credential previously saved for performing authenticated LLM Queries...

CVE-2025-62327

The CVE-2025-62327 affects HCL DevOps Deploy versions 8.1.2.0 through 8.1.2.3. A user with LLM configuration privileges may recover credentials saved for performing authenticated LLM Queries, indicating improper access control around LLM credentials. Root cause described across sources is insuffi...

LLMs, You Can Evaluate It! Design of Multi-Perspective Report Evaluation for Security Operation Centers

Security operation centers SOCs often produce analysis reports on security incidents, and large language models LLMs will likely be used for this task in the near future. We postulate that a better understanding of how veteran analysts evaluate reports, including their feedback, can help produce...

Exploit for CVE-2025-68664

--- 📑 Table of Contents - 🎯 Executive Summary-executive...

LLM-Driven Feature-Level Adversarial Attacks on Android Malware Detectors

The rapid growth in both the scale and complexity of Android malware has driven the widespread adoption of machine learning ML techniques for scalable and accurate malware detection. Despite their effectiveness, these models remain vulnerable to adversarial attacks that introduce carefully crafte...