83 matches found
masq
Sit where the model sits. See what it can see. MASQ MASQ is a local security CLI for AI agent stacks. It sits in the same seat the model sits in: it reads MCP tool catalogs, completes the MCP handshake, looks through session logs, and probes local model HTTP ports. Then it prints findings and...
VulnLLM-R
VulnLLM-R: Specialized Reasoning LLM for Vulnerability Detection Paper: arXiv:2512.07533 Code & Data: GitHub Demo: Web demo Model: 7B Model Environment and dataset 🛠️ Create environment Install Git LFS and clone the repository LFS files are fetched automatically during clone git lfs install...
deleting-the-trace
Control-token injection attacks on tool-using agents Code and logged measurements for a study of two input-level attacks on tool-using language model agents. The first attack appends a short string of a model's own channel-control tokens to untrusted input; the tokenizer reads it as an...
Vulnerabilities fixed in VMware products
VMware has patched vulnerabilities in the VMware vCenter Directory Service, VMware Telco Cloud Platform, VMware SDDC Manager, and VMware Tools for Windows. These vulnerabilities affect multiple VMware products and versions. In the VMware vCenter Directory Service and VMware Telco Cloud Platform,...
Fedora 44 : perl-Imager (2026-9754401c44)
The remote Fedora 44 host has a package installed that is affected by multiple vulnerabilities as referenced in the FEDORA-2026-9754401c44 advisory. 6 CVEs fixed across versions 1.0371.041: - CVE-2026-124720 TGA memory leak - CVE-2026-124721 PNG file limit checks - CVE-2026-124722 PNM file limit...
llm-censorship-steering
LLM Censorship Steering This repository contains code implementation for Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control by Hannah Cyberey and David Evans. We introduce a method that finds "steering vectors" from LLM internals for detecting and controlling the...
Linux Distros Unpatched Vulnerability : CVE-2026-99475
The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - genirq: Defer kfreeirqaction until irqthreaddtor finishes When an interrupt thread exits abnormally via doexit e.g. the interrupt handler oopses without returni...
Capable yet Parsimonious: Extracting and Characterizing Hidden Chain-Of-Thought in Frontier Models
The rapid capability gains of frontier language models are widely attributed to improved reasoning abilities, yet this cannot be verified as raw CoT traces in closed-source systems are hidden. By registering a simple custom tool through a standard API feature, we induce frontier models to...
U.S. Agencies Accuse China AI Firms of Distilling Claude, GPT, Gemini, and Grok
U.S. cybersecurity and intelligence agencies have accused China-based artificial intelligence AI companies of conducting "systematic extraction" of proprietary functionalities and capabilities of American frontier models through distillation attacks. The activity has been described as occurring a...
Malicious code in deep-thought-relay-client (npm)
The package deep-thought-relay-client was found to contain malicious code. --- Credit: OpenSSF source...
The Implications of Linguistic Illegibility for LLM Security
LLMs are trained to generate natural language. However, various strands of evidence indicate that an LLM's externalized linguistic outputs and mechanistically-extracted linguistic features can be an unreliable lens for understanding internal model computation. We introduce the term "linguistic...
Stealing Reasoning Traces from Proprietary LLM APIs
Leading large language model providers now conceal their models' step-by-step reasoning, or chain-of-thought, to protect intellectual property and limit information leakage. Rather than storing these traces server-side, providers return them to the client as blocks of encrypted text, which the...
Paged Out! Magazine Issue 6
Paged Out! magazine issue 6 includes sections related to art, artificial intelligence, assembly, cryptography, demo scene, file formats, food for thought, game development, hardware, networks, OS internals, programming, retro stuff, reverse engineering, and security / hacking...
Evading Chain-Of-Thought Monitoring through Model Poisoning
Chain-of-thought CoT monitoring is an increasingly important component of AI safety stacks but relies on the assumption that a model's reasoning trace is informative about its actions. This work studies the limits of CoT monitoring through the lens of model poisoning. We demonstrate that backdoor...
Cybersecurity Detection Classification with Reasoning-Enabled Language Models
A major issue in Security Operations Centers SOCs is alert fatigue, as the number of detections reported is more than staff can triage in a given day. Prior work prompts or fine-tunes large language models LLMs to emit a triage label directly, but does not train them to reason about whether a...
PYSEC-2026-2645 MetaGPT has an eval injection in metagpt/strategy/tot.py
A vulnerability was identified in FoundationAgents MetaGPT up to 0.8.2. This affects the function generatethoughts of the file metagpt/strategy/tot.py of the component Tree-of-Thought Solver. The manipulation leads to code injection. It is possible to initiate the attack remotely. The exploit is...
MetaGPT has an eval injection in metagpt/strategy/tot.py
A vulnerability was identified in FoundationAgents MetaGPT up to 0.8.2. This affects the function generatethoughts of the file metagpt/strategy/tot.py of the component Tree-of-Thought Solver. The manipulation leads to code injection. It is possible to initiate the attack remotely. The exploit is...
GDM AI Control Roadmap
AI agents are rapidly accelerating work at frontier AI companies, helping with AI R&D, cyber-defence, and advancing scientific discoveries. As these agents become more tightly integrated into our systems, unlocking their full potential requires rethinking how we do security. We should not assume...
PT-2026-59295
A vulnerability was identified in FoundationAgents MetaGPT up to 0.8.2. This affects the function generate thoughts of the file metagpt/strategy/tot.py of the component Tree-of-Thought Solver. The manipulation leads to code injection. It is possible to initiate the attack remotely. The exploit is...
Reasoning As an Attack Surface: Adaptive Evolutionary CoT Jailbreaks for LLMs
Large Reasoning Models LRMs have demonstrated remarkable capabilities in reasoning and generation tasks and are increasingly deployed in real-world applications. However, their explicit chain-of-thought CoT mechanism introduces new security risks, making them particularly vulnerable to jailbreak...