Astra Linux – Vulnerabilities in Linux, Linux-5.10, Linux-5.15, Linux-6.1

In the Linux kernel, the following vulnerabilities have been resolved: netfilter: nftables: The backend for setting up DEAD bits was changed to use the GC transaction API. The old and buggy gc API and the busy mark approach have been replaced with the GC transaction API. No set elements are remov...

Astra Linux – Vulnerability in Linux 5.10

In the Linux kernel, the following vulnerability has been resolved: ext4: Fixed a traversal bug in ext4mbusepreallocated. During allocation, when searching for pre-allocations PA in the per-inode rbtree, we cannot perform a direct traversal of the tree because ext4mbdiscardgrouppreallocation may...

ROOT-APP-PYPI-CVE-2026-32274 CVE-2026-32274 in rootio-black - Patched by Root

Root has patched CVE-2026-32274 in the rootio-black package for Root:PyPI. Multiple fixed versions available...

Assessing Automated Prompt Injection Attacks in Agentic Environments

Indirect prompt injection poses a critical threat to LLM agents that interact with untrusted external data, yet automated attack methods--proven effective for jailbreaking--remain underexplored in realistic agentic settings. We present a comprehensive empirical evaluation of automated prompt...

MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents

External memory has become a core component of modern web agents, enabling long-horizon reasoning through the retrieval of past experiences. However, this paradigm introduces a critical vulnerability: malicious content injected into memory can be persistently recalled and repeatedly influence age...

PT-2026-47061

Excited to share my research was accepted at @BlackHatEvents USA 2026! 🎩 I'll present how I achieved interactive access to users' AI assistants by chaining: 🔓 Prompt injection 🔓 Privilege escalation 🔓 Path traversal 🔓 .toml injection 🔓 and finally an LD PRELOAD exploit The impact: 🚨 CVE-2026-3219...

RedEdit: Agentic Red-Teaming of Image Safety Classifiers Via MCTS-Guided Photo-Editing

Image safety classifiers serve as a critical component of contemporary content moderation systems on the internet. However, their resilience against user-style malicious image editing remains underexplored. Such behaviors are highly prevalent in daily scenarios but difficult to fully reproduce. T...

CVE-2026-37235

FlexRIC v2.0.0 trusts the xappid field from E42 message payloads without binding it to the sender's SCTP association. The validation function validxappid only checks that the value is within the assigned range. A remote unauthenticated attacker can impersonate any xApp by specifying their xappid ...

CVE-2026-37235

FlexRIC v2.0.0 trusts the xapp_id field from E42 payloads without binding it to the sender’s SCTP association. The valid_xapp_id() check only ensures the value is within the assigned range, enabling a remote unauthenticated attacker to impersonate any xApp by supplying their xapp_id in requests t...

MaskForge: Structure-Aware Adaptive Attacks for Jailbreaking Diffusion Large Language Models

Diffusion large language models dLLMs generate text by iteratively denoising partially masked sequences under bidirectional context, exposing a safety surface distinct from autoregressive LLMs. Because mask tokens are native inputs and tokens are committed by confidence rather than position,...

WordPress Media Library Assistant plugin <= 3.35 - Cross-Site Request Forgery vulnerability

Cross-Site Request Forgery vulnerability discovered by Jack Pas Dark. - Black Lantern Security in WordPress Plugin Media LIbrary Assistant versions = 3.35...

Evolving Skill-Structured Attack Memory Enhances LLM Jailbreaking

Jailbreak attacks on large language models LLMs aim to induce LLMs to produce content that they are expected to refuse. Automated black-box jailbreak generation is especially important for safety evaluation, where the attacker observes only model outputs and needs to automatically search for...

MRMMIA: Membership Inference Attacks on Memory in Chat Agents

Membership inference attacks MIAs test whether a target data record belongs to a system's private data, and have become a standard tool to measure privacy leakage in machine learning systems. Prior work has primarily focused on training corpora or retrieval databases. However, MIAs against agent...

VibeHacking

👾 Welcome to Vibe Hacking By BlackPC, Vine & Foxxino Inc...

Babel: Jailbreaking Safety Attention Via Obfuscation Distribution Optimized Sampling

Despite rigorous safety alignment, Large Language Models LLMs remain vulnerable to jailbreak attacks. Existing black-box methods often rely on heuristic templates or exhaustive trials, lacking mechanistic interpretability and query efficiency. In this study, we investigate an intrinsic...

Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw

Agentic language-model systems increasingly rely on mutable execution contexts, including files, memory, tools, skills, and auxiliary artifacts, creating security risks beyond explicit user prompts. This paper presents DeepTrap, an automated framework for discovering contextual vulnerabilities in...

Re-Triggering Safeguards within LLMs for Jailbreak Detection

This paper proposes a jailbreaking prompt detection method for large language models LLMs to defend against jailbreak attacks. Although recent LLMs are equipped with built-in safeguards, it remains possible to craft jailbreaking prompts that bypass them. We argue that such jailbreaking prompts ar...

CVE-2026-41511

OpenMcdf is a fully .NET / C library to manipulate Compound File Binary File Format files, also known as Structured Storage. Prior to version 3.1.3, OpenMcdf does not detect cycles in the directory entry red-black tree of a Compound File Binary CFB document. A crafted CFB file with a cycle in the...

OpenMcdf 安全漏洞

OpenMcdf is an open-source .NET library for manipulating compound file binary formats. Versions of OpenMcdf prior to 3.1.3 contained security vulnerabilities, which stemmed from undetected loops in the red-black tree used to manage directory entries, potentially leading to infinite loops...

Black_Box-Penetration-Testing

BlackBox-Penetration-Testing Black-box penetration test again...