llm-circuit-finder

llm-circuit-finder I replicated Ng's RYS method and found that duplicating 3 specific layers in Qwen2.5-32B boosts reasoning by 17% and duplicating layers 12-14 in Devstral-24B improves logical deduction from 0.22→0.76 on BBH — no training, no weight changes, just routing hidden states through th...

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...

aixcc-afc-atlantis

Atlantis This repository contains Atlantis , Team Atlanta's Cyber Reasoning System CRS submitted to the AIxCC Final Competition AFC. Atlantis is designed to automatically discover and analyze vulnerabilities in software systems. The main system implementation is located in the...

SecFlow

SecFlow Fully Automated Multi-Vector Threat Analysis Pipeline Drop in any file, URL, IP, domain, or image. SecFlow routes it through specialized analyzers, reasons about findings with AI, and produces a professional security report with YARA rules, SIGMA rules, and exportable PDF — automatically...

AutoRAN-public

🧠 AutoRAN: Automated Hijacking of Safety Reasoning in Large Reasoning Models AutoRAN is an automated Hijacking of Safety Reasoning that leverages less-aligned secondary auxiliary models to simulate reasoning traces, generate narrative prompts, and iteratively refine those prompts to bypass safety...

vulnhawk

AI-powered code security scanner that finds vulnerabilities Semgrep and CodeQL miss. Quick Start • GitHub Action • Comparison • Languages • FAQ The Problem Traditional SAST tools rely on pattern matching and AST rules. They excel at catching known vulnerability patterns, but they fundamentally...

GuardReasoner-VL

GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning Yue Liu, Shengfang Zhai, Mingzhe Du Yulin Chen, Tri Cao, Hongcheng Gao, Cheng Wang Xinfeng Li, Kun Wang, Junfeng Fang, Jiaheng Zhang, Bryan Hooi 1National University of Singapore, 2Nanyang Technological University To enhance the safety ...

sgxray

SGXRay: Automating Vulnerability Detection for SGX Apps Intel SGX protects isolated application logic and sensitive data inside an enclave with hardware-based memory encryption. To use such hardware-based security mechanism requires a strict programming model on memory usage， with complex APIs in...

Agentic-SOC-Simulation

🛡️ Agentic SOC Parallel Simulation ASPS: Autonomous Security Operations Parallel Simulation Center Redefining the Future of Security Operations: Full-Stack SOC Parallel Simulation, Multi-Agent Collaboration, and Data-Driven Autonomous Defense 🇨🇳 中文文档 Chinese Version This project is a cutting-edge...

OdTM

OWASP Ontology-driven threat modelling OdTM framework Ontology-driven Threat Modelling OdTM framework is a set of means for implementation of an ontological approach into automatic threat modelling of computer systems. It is an OWASP Incubator Project. The ontological approach, provided by the Od...

Sentinel-GPT

Sentinel-GPT: Agentic AI for Threat Hunting 🛡️ An automated cybersecurity pipeline that transforms raw Azure logs into actionable intelligence using LLM-based reasoning. 📖 Overview Sentinel-GPT is an AI Agent specialised for threat hunting in cybersecurity. It is integrated with Azure Log Analytic...

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...

codeql-workshop-cve-2021-21380

CodeQL workshop for Java: Finding a SQL injection In this workshop we will use syntactical and semantic reasoning to find a SQL injection in the XWiki platform's rating component documented by CVE-2021-21380 Contents CodeQL workshop for Java: Finding a SQL injection Contents Prerequisites and set...

NetLogic

NetLogic Cloud-Native Attack Surface Mapper & Vulnerability Correlator — v3.0 NetLogic is a network security platform combining active port scanning, CVE correlation live NVD API, SSL/TLS analysis, HTTP security auditing, DNS/email security assessment, subdomain takeover detection, passive OSINT,...

KidnapRAG

KidnapRAG: A Black-Box Attack for Hijacking Reasoning in Agentic Retrieval-Augmented Generation Systems 🎓 Paper | 📄 Datasets | 🚀 Quick Start Quick Start Run the following commands from the parent directory of the cloned KidnapRAG repository. cd KidnapRAG conda create -n KidnapRAG python=3.10 cond...

claude-XML-injection

Claude Sonnet 4.6 — Credential Fabrication via XML Tag Injection Reported: June 14, 2026 | Patched: June 20, 2026 | Anthropic Response: None 56 days Researcher: X1NON Affected Model: Claude Sonnet 4.6 and other non-Haiku Claude models Severity: High CVSS 8.7 Status: Patched — No bounty awarded, n...

Where Did the Repair First Go Wrong? Localizing the Origins of Silent Failures in Agentic Vulnerability Repair

Localizing where an LLM-based agent first fails to uphold security during a repair can show which stage of its workflow needs an additional safeguard. This is difficult for silent failures, which are patches that pass syntactic and functional checks but still contain a security vulnerability...

Correct Verdicts, Flawed Reasoning: Structured Auditing of LLM-Based Vulnerability Reasoning

Large Language Models LLMs are increasingly deployed for automated software vulnerability analysis. Binary classification alone is insufficient; practitioners need explanations to triage bugs and engineer patches. Standard practice relies on Chain-of-Thought CoT prompting, but free-form reasoning...

Defense-In-Depth at the Perception-Reasoning Interface of LLM-Centric Agentic UAV Swarms

Large Language Models LLMs increasingly support Uncrewed Aerial Vehicle UAV swarm operations such as data collection scheduling, where the model reads structured sensor reports and decides which sensors to visit. An adversary who quietly manipulates those reports can redirect the swarm without...

OpenAI Disrupts Reasoning Extraction Campaign Linked to Moonshot AI Associates

OpenAI on Wednesday said it identified and disrupted a coordinated distillation campaign that was designed to illicitly extract protected reasoning from its artificial intelligence AI models. A "core cluster of the activity," going back to the first week of July, has been attributed to individual...