Configuration, Not Conscience: A Large-Scale Empirical Study of LLM System Prompts

Leaked system prompts are often treated as windows into the hidden values of commercial language models, yet their composition is rarely studied at scale. We analyze a merged corpus of 407 leaked, reconstructed, or officially published system prompts from 62 vendors across four community...

MetaPermit: Scalable and Auditable Access Control for AI Agents Via LLM-Inferred Meta-Attributes

The rise of autonomous AI agents equipped with tools has introduced significant security risks, ranging from unintended tool misuse to adversarial manipulation through Indirect Prompt Injection IPI attacks. In practice, deployed agent systems such as OpenAI Codex and Claude Code protect tool...

TempQ-Jail: Query-Constrained Candidate Ranking for Text-To-Video Jailbreak Attacks

Existing text-to-video T2V jailbreak methods mainly seek more effective or stealthier attack candidates. In guarded T2V systems, however, video generation and security evaluation are costly, so an attacker often cannot test a large candidate pool. We therefore formulate T2V jailbreak as a...

Breaking the Black Box: Byte-Level Boundary Inference of Real-World Antivirus Systems

Whitepaper called Breaking The Black Box: Byte-Level Boundary Inference Of Real-World Antivirus Systems...

GitHub Engagement Signals for CVE Prioritization: The GitHub Popularity Metric (GPM)

Whitepaper called GitHub Engagement Signals For CVE Prioritization: The GitHub Popularity Metric GPM...

Weaponizing Ground Truth: Data Poisoning Attacks by Exploiting Boundary Misalignment between Antivirus Software and Learning-Based Detectors

Machine-learning ML-based malware detectors are commonly trained using labels obtained from antivirus AV engines and aggregation services e.g., VirusTotal. This practice assumes AV-generated labels provide reliable supervision. However, small byte-level modifications can substantially alter AV...

FARE: Forensic Acceptance Region Estimation for Catching Bait-And-Switch Image Generators

Modern AI image generators are increasingly deployed as opaque APIs, where customers can query the deployed service, but cannot inspect model weights or architecture. This creates a practical challenge: a provider may pass governance certification with one generator and later silently switch to a...

FeatMark: Feature-Level Watermark Protection against Mimicry Attacks with Diffusion Models

Text-to-image diffusion models enable data-efficient "mimicry" attacks, wherein adversaries fine-tune the model on a handful of public photos to synthesize convincing forgeries of a target individual. A common countermeasure is to embed imperceptible, low-energy watermarks, yet recent studies sho...

How to Break the Miranda Signature Scheme over Matrix Gabidulin Codes

The Miranda signature scheme relies on masking a matrix code which disposes of a masked underlying structure, and the knowledge of which allows for efficient error decoding. We consider a Gabidulin code which is expanded into a matrix code, that is only \Fq-linear. An additional masking is then...

AGATE: Provenance-Based Runtime Defense against Compositional Attacks on LLM Agents

LLM agents can produce harmful effects through sequences of ordinary operations. Judging such actions requires establishing both the authority that permits them and the origin of the data they carry. We present AGATE, an authorization and data-provenance gate at instrumented agent-harness...

Crypto-Bound Identity-Verified Capability Tokens for Coordinating Distributed AI Agents: A Proposal

The prospect of fully autonomous transactional agents did not appear on the horizon until the advent of high capability language models. With such models, the operational benefits of adaptive task orchestration and independent but constrained decision making are tantalizing for enterprises and...

XPhysICS: Cross-Physical-Domain Threat Grounding for Industrial Control Systems Security

Industrial control system ICS threats documented for one plant can express cyber-physical effects relevant to another, but semantic similarity alone does not establish whether those effects are structurally admissible or evaluable on a target. We present XPhysICS, a provenance-aware,...

Frame the Adversary: A Structure-Aware Attack Methodology

Frequency-based adversarial attacks have recently grown popular by exploiting spectral sensitivities shared across neural architectures. Unlike spatial perturbations, frequency-based attacks expose deeper vulnerabilities, making them especially valuable for robust evaluation of safety-critical an...

Context-Aware Functional Modeling for Android Third-Party Library Detection

Third-party libraries TPLs are widely used in Android apps, but their reuse can introduce security risks and interfere with downstream program analyses. Existing Android TPL detection approaches face two key limitations: their hand-crafted features are fragile under aggressive code transformation...

Fast and Secure Simultaneous Authentication of Equals for WPA3

The Simultaneous Authentication of Equals SAE protocol, introduced in WPA3, provides robust protection against offline dictionary attacks against a network Pre-Shared Key PSK and also protection of session keys from other people knowing that PSK. However, its high computational cost for the...

JevAdvBench: A Benchmark and Black-Box Attacks for Reinforcement Learning for Calibrated Decisions Models

Models trained with reinforcement learning for calibrated decisions RLCD, such as Jev, answer a typed question about an input, the state, with a probability, a choice, or a score, and software acts on the answer without a person reading it. Their robustness has not been measured: adversarial...

GNU Privacy Guard 2.5.24

GnuPG the GNU Privacy Guard or GPG is GNU's tool for secure communication and data storage. It can be used to encrypt data and to create digital signatures. It includes an advanced key management facility and is compliant with the proposed OpenPGP Internet standard as described in RFC2440. As suc...

Improving the Reliability of Anomaly Detection for Encrypted OPC UA Traffic over Private 5G

Open Platform Communications Unified Architecture OPC UA is increasingly deployed over private 5G networks in industrial environments, where end-to-end encryption prevents payload inspection by network-based intrusion detection systems IDSs. Although payload-agnostic statistical features extracte...

Automated Abstraction Refinement for Information Flow Security in Embedded Systems

Information flow analysis IFA is a powerful technique for verifying confidentiality and integrity and is therefore highly desirable for security-sensitive embedded systems. However, as these systems are inherently concurrent and time-dependent, existing IFA for embedded systems tend to be either...

CISA: Recognizing and Addressing Threats to Statewide Voter Registration Databases

Recently declassified records revealed that China breached multiple state voter registration systems prior to the 2020 election. Unfortunately, this is not a new phenomenon. Unclassified reports from the last decade reveal that Statewide Voter Registration Databases VRDB are attractive targets fo...

Secure Polarization-Shift Backscatter Identification Applied to Battery-Free BLE Sensors Powered by Wireless Power Transfer

This paper presents a lightweight and protocolindependent security mechanism for battery-free Bluetooth Low Energy BLE sensor nodes operating in Simultaneous Wireless Information and Power Transfer SWIPT architecture. The proposed approach exploits polarization-shift backscattering of the wireles...

The Vulnerability of Neural Audio Watermarks under Speech Enhancement

Neural audio watermarks are increasingly deployed in commercial speech generation systems to make AI-generated speech traceable, yet their robustness has been studied mainly under conventional signal distortions. Since a watermark can be regarded as imperceptible noise added to the speech signal,...

ClaimMirage: When Self-Claims in Domain Names Change LLM Threat Judgments

Short claims such as not-phishing or official can change how a large language model LLM judges a domain name, without explicit prompt-injection commands. We study this manipulation as ClaimMirage: a name under inspection claims its own safety or approval. We analyze 622,080 judgments across 64...

Wireshark Analyzer 4.6.9

Wireshark is a GTK+-based network protocol analyzer that lets you capture and interactively browse the contents of network frames. The goal of the project is to create a commercial-quality analyzer for Unix and Win32 and to give Wireshark features that are missing from closed-source sniffers. Thi...

REDCap 17.4.3 Survey Passthru Data Import Checker

REDCap versions 13.3.0 through 16.0.48, 17.0.0 through 17.3.9, and 17.4.0 through 17.4.3 contain an unauthenticated remote code execution vulnerability in survey passthrough routing and Data Import processing. This tool fingerprints instances and probes public-survey passthrough routes...

Too Late to Slash: Coordinating a Risk-Free Equivocation Attack

Slashing is commonly argued to secure proof-of-stake blockchains by confiscating the stake of misbehaving validators. The usual justification is that, without slashing, validators can solicit an equivocation attack by signing conflicting blocks: if enough others join, the attack succeeds;...

Input-Layer Starvation: Why Per-Layer Pruning Breaks IoT Intrusion Detectors

Intrusion detectors for small Internet-of-Things IoT devices are usually compressed by pruning and judged by overall accuracy. We show that this hides a severe class-level failure, find its cause, and give low-overhead prevention and repair. On CICIoT2023, a two-layer convolutional detector prune...

A Large-Scale Empirical Study of Modern Phishing Email Content

Phishing remains one of the most pervasive threats to Internet users, and email remains its predominant delivery channel. Email content is the attack surface of phishing: it is what the victim reads and what automated defenses inspect. Yet the composition of modern phishing content is poorly...

Prompt Injection Detection for Email Agents through Attack Chain Modeling

Large language model email assistants are particularly vulnerable to indirect prompt injection because untrusted email content can be retrieved into the model context and influence subsequent tool use. Existing prompt injection detectors mainly formulate this problem as binary malicious text...

Joern 4.0.636

Joern is the bug hunter's workbench. With this tool, you can uncover attack surface, sloppy coding practices, and variants of known vulnerabilities using an interactive code analysis shell. Joern supports C, C++, LLVM bitcode, x86 binaries via Ghidra, JVM bytecode via Soot, and Javascript...

TOR Virtual Network Tunneling Tool 0.4.9.13

Tor is a network of virtual tunnels that allows people and groups to improve their privacy and security on the Internet. It also enables software developers to create new communication tools with built-in privacy features. It provides the foundation for a range of applications that allow...

ASRock IO Driver version 1.00.00.0000 Improper Access Control

ASRock IO Driver version 1.00.00.0000 contains an improper access control vulnerability in AsrDrv103.sys. Included research analyzes the vulnerable driver and its encrypted IOCTL mechanism...

CISA: 2026 Election Infrastructure Security Plan

Election infrastructure requires steadfast security and resilience. It is a vital national interest and one of the Department of Homeland Security’s highest priorities to support the greater national security landscape. The American people’s confidence that their vote will be counted as cast is...

CISA: CISA Election Report

From around 2019 to 2024, the Cybersecurity and Infrastructure Security Agency CISA conducted a set of technical and operational activities to evaluate the security of certain U.S. election systems. These activities, all carried out upon the request of system owners and operators, included direct...

LLM Agents Can Easily Tamper with Their Own Traces

Asynchronous monitoring, incident investigations, and compliance audits primarily rely on agent traces to reconstruct what happened. These analyses assume that LLM agents cannot tamper with their own execution traces. We show that local LLM agents such as Claude Code, Codex, Antigravity, Open Cod...

Instrumental Monitor Evasion Emerges under Ordinary Task Pressure

A central concern in AI safety is that agents may treat oversight as an obstacle when it conflicts with completing their goals. We study instrumental evasion, the propensity of LLM agents to circumvent runtime monitoring as a means of completing ordinary tasks. We introduce EvasionBench, a...

A Data-Driven Analysis of Infostealer Malware Victims

Infostealer malware infects devices worldwide and harvests their most sensitive contents: credentials, browser sessions, private keys, and access certificates. Yet its impact on victims remains difficult to study without an ethical, legal, and curated research dataset. To close this gap, we build...

Trusted Model Environment for Private Semantic Computations

A private semantic computation primitive enables parties to privately compute over structured and unstructured data that requires understanding its semantics, context, and relationships. Standard cryptographic primitives e.g., multiparty computation do not readily support such computation...

ENDOPROMPT: Victim-Side Pseudo-References for Utility Degradation

Prompt injection can degrade benign task performance without eliciting harmful content. Yet many attack objectives depend on task labels or predefined target responses. We present ENDOPROMPT, a white-box method that learns utility-degrading prefixes from unlabeled instructions. Its generator take...

Computational Cryptography from Pseudoentanglement

The advent of pseudoentanglement and computational entanglement theory bootstrapped a wave of research at the intersection of computer science and information theory. In parallel, computational cryptography has undergone substantial development, prompted by the introduction of pseudorandom states...

Template Ageing and Longitudinal Verification in Fixed-Text Keystroke Dynamics: A Subject-Disjoint Study across Eight Weeks

Behavioural biometric templates are widely believed to degrade as the gap between enrolment and verification grows, but few studies measure this template ageing effect directly under controlled conditions. We collected a longitudinal dataset of 40 fixed passwords, each typed four times per weekly...

Hard Stop: Kernel-Level Preemption and Containment for Rogue Agentic Execution

In July 2026, an unconstrained autonomous agent participating in a frontier AI cybersecurity evaluation harness breached its evaluation sandbox, established an external command-and-control foothold, and executed a multi-stage intrusion into Hugging Face's production multi-tenant dataset conversio...

OllamaDrama: Designing and Deploying a Honeypot to Measure Attacks on Exposed LLM Infrastructure

Publicly exposed large language model LLM infrastructure creates a growing attack surface, yet real-world targeting remains poorly understood. We present Ollure, a low- and medium-interaction honeypot that emulates the Ollama API without a backend LLM. Spanning four deployments across cloud and...

AEGIS: Audio Endogenous Guarding Via Internal Signals against Large Audio-Language Model Jailbreaks

Large audio-language models LALMs expand language models to process and interpret audio, but also expose them to heterogeneous audio jailbreaks. We ask whether successful jailbreaks reflect failures to recognize harmful intent or failures occurring after such recognition. Layer-wise probing revea...

Beyond Centralized Policy Decision Points: Decentralized Sticky Policy Authorization through Evidence Quorums

Sticky policies remain attached to protected data so that access restrictions can persist across systems, yet request-time authorization often still depends on a centralized Trusted Authority or Policy Decision Point PDP. This paper presents SPEAR-Q, a decentralized framework in which independent...

TP-CRIV: A Framework for Third-Party Challenge-Response Identity Verification of AI Models

Artificial intelligence AI models are increasingly deployed through remote services, making model misappropriation a growing concern. Existing approaches, including watermarking, fingerprinting, and model similarity analysis, primarily rely on predefined evidence or direct behavioral comparison a...

Security Limits of Mining Before Validation in Nakamoto Consensus

Mining before validation allows miners to extend a newly received block before completing its validity checks, giving them a head start in the race for the next block reward. This head start, however, comes with a security risk: rejecting one invalid block also discards the honest work built on i...

From WPT to Encrypted Telemetry: A Battery-Free Backscattering-Based Polarimetric Wireless Sensor

This work introduces an indoor Battery-Free Wireless Sensing Node powered through radiative Wireless Power Transfer WPT. The proposed platform targets secure, energyefficient active sensing and overcomes key limitations of many prior battery-free approaches, which commonly provide neither on-node...

The Fly That Stopped: Mushroom-Body-Inspired Habituation As a Reward-Free Scheduling Prior for Autonomous Penetration Testing

Autonomous security-testing agents can spend much of a fixed action budget repeating earlier tool selections. We evaluate a reward-free scheduler inspired by mushroom-body novelty processing in Drosophila. It combines sparse state encoding with decaying habituation counters over structural URL...

TraceGuard: Adaptive Multimodal Poison Filtering through Cross-Feature Rank Agreement

Multimodal training relies on image-text corpora collected from external sources, creating opportunities for attackers to poison the data. Stealthy attacks can preserve plausible image-text pairs while concealing the differences used by detectors, so apparently clean data can still redirect the...