4824 matches found
Auditing Anonymous AI Models: A Four-Stage Protocol for Black-Box Identity Verification
The 2025--2026 AI market has seen a wave of stealth releases: frontier models launched anonymously on developer platforms under codenames. For their users, identity determines data-handling terms, supply-chain risk, and capability expectations. No validated methodology exists for black-box identi...
WuppieFuzz v1.7.1
WuppieFuzz v1.7.1 TNO developed WuppieFuzz, a coverage-guided REST API fuzzer developed on top of LibAFL, targeting a wide audience of end-users, with a strong focus on ease-of-use, explainability of the discovered flaws and modularity. WuppieFuzz supports all three settings of testing black box,...
SIR: Self-Improving Red-Teaming for Compute Use Agents
Computer use agents CUAs are vision-language models that perceive a screen and act on a real operating system through mouse, keyboard, and terminal, and they are increasingly deployed to automate everyday digital tasks. Because they can be exposed to untrusted content while operating, they are...
linux-server-pentest-lab
🔐 Linux Black-Box-Style Penetration Test — Full Attack Chain...
REPLICANT: Learning Policies for Evading and Hardening Malware Detectors
To determine the real-world effectiveness of machine learning based malware detection, it is vital to evaluate its robustness against highly capable adversaries. However, state-of-the-art attacks do not effectively model realistic adversaries, as they often assume access to privileged information...
htb-labs-orion
🚀 HTB Labs - Orion English | EspañolREADME-ES...
RedEvoAgent: Automatic Red-Teaming Agent with Experience-Driven Skill Evolution
LLM-based agents are increasingly deployed in product-level execution harnesses, where jailbreaks can trigger harmful tool use and persistent state changes, creating greater risks than unsafe text generation alone. Existing automatic red-teaming methods often rely on fixed attacks, while recent...
Daydreaming: Stealing Hidden Agent Skills through Black-Box Task Interaction
Agent skills bundle instructions, reference data, and executable helpers that let a general agent perform specialized tasks. Hosted providers can keep these files secret while selling access to task results, making the skill itself a valuable target. Existing disclosure defenses can block request...
PT-2026-81978
The Betheme theme for WordPress is vulnerable to Stored Cross-Site Scripting via the theme's 'icon box 2' shortcode in all versions up to, and including, 28.4 due to insufficient input sanitization and output escaping on user supplied attributes. This makes it possible for authenticated attackers...
Unity Linux 20.1050e / 20.1060e / 20.1070e Security Update: ffmpeg (UTSA-2026-104785)
The Unity Linux 20 host has a package installed that is affected by a vulnerability as referenced in the UTSA-2026-104785 advisory. FFmpeg before commit acf5d7c contains a heap buffer overflow in the hvcC box writer. When writing an HEVC configuration record with more NAL units of a single type...
A Self-Evolving Multi-Agent Framework Defense against LLM Jailbreak Attacks
Large language models LLMs remain vulnerable to jailbreak attacks that exploit techniques such as role-playing, obfuscation, code transformation, and multi-step indirection to elicit harmful outputs. As jailbreak strategies keep emerging, defenses have proliferated in an ongoing cat-and-mouse gam...
NeuronFuzz: Safety Neuron Guided Fuzzing for LLM Safety Evaluation
Safety evaluation is critical for assessing whether aligned Large Language Models LLMs remain robust against jailbreak attacks. Existing automated testing methods, however, largely rely on response-level feedback: each candidate prompt typically requires generating a target-model response to...
CVE-2026-75019
The Cozy Blocks – Page Builder for Gutenberg Editor & FSE with 700+ Patterns, 58 Blocks & Templates plugin for WordPress is vulnerable to Stored Cross-Site Scripting via cozyHoverEffect Block Attribute in all versions up to, and including, 2.2.16 due to insufficient input sanitization and output...
cyber-apocalypse-2026-salt-crown
Cyber Apocalypse CTF 2026: The Salt Crown — Writeups Writeups...
SUSE CVE-2026-75141
FFmpeg before commit acf5d7c contains a heap buffer overflow in the hvcC box writer. When writing an HEVC configuration record with more NAL units of a single type than the count field can represent, the NAL unit count overflows, causing a heap buffer overflow. A crafted HEVC input file triggers...
PT-2026-81879
Name of the Vulnerable Software and Affected Versions SililaWijesinghe Food Ordering System versions up to ba314e897e3365600461e5ea59432e39ceaa0fa5 Description An issue exists in the /search.php endpoint where manipulation of the search box variable allows for SQL injection. This allows remote...
Malicious code in ultra-safe-box (RubyGems)
--- -= Per source details. Do not edit below this line.=-...
MAL-2026-15804 Malicious code in ultra-safe-box (RubyGems)
--- -= Per source details. Do not edit below this line.=-...