Forwarded from CyberSecurityTechnologies (-CST-)
BSF.pdf
658.6 KB
#tools
#Research
#Malware_analysis
"Investigating the Effectiveness of Bayesian Spam Filters in Detecting LLM-modified Spam Mails", 2024.
]-> Pipeline for Testing Spam Filter's Effectiveness against LLM-modified Spam Mails:
https://github.com/MalteJosten/llm-spam-test
#Research
#Malware_analysis
"Investigating the Effectiveness of Bayesian Spam Filters in Detecting LLM-modified Spam Mails", 2024.
]-> Pipeline for Testing Spam Filter's Effectiveness against LLM-modified Spam Mails:
https://github.com/MalteJosten/llm-spam-test
Forwarded from CyberSecurityTechnologies (-CST-)
RAGent.pdf
1.1 MB
#tools
#Research
#Blue_Team_Techniques
"RAGent: Retrieval-based Access Control Policy Generation", 2024.
]-> https://github.com/accessframework/RAGent
#Research
#Blue_Team_Techniques
"RAGent: Retrieval-based Access Control Policy Generation", 2024.
]-> https://github.com/accessframework/RAGent
Forwarded from CyberSecurityTechnologies (-CST-)
AI_Safety_RT.pdf
2.5 MB
Forwarded from CyberSecurityTechnologies
AIJack.pdf
1001.6 KB
#tools
#MLSecOps
"Stealth by Conformity: Evading Robust Aggregation through Adaptive Poisoning", 2025.
]-> AIJack: Security and Privacy Risk Simulator for Machine Learning
// In this paper, we challenge this underlying assumption by showing that a model can be poisoned while keeping malicious updates within the main distribution
#MLSecOps
"Stealth by Conformity: Evading Robust Aggregation through Adaptive Poisoning", 2025.
]-> AIJack: Security and Privacy Risk Simulator for Machine Learning
// In this paper, we challenge this underlying assumption by showing that a model can be poisoned while keeping malicious updates within the main distribution
🔥3
Forwarded from CyberSecurityTechnologies
#tools
#AIOps
#MLSecOps
Same Model, Different Hat:
Bypassing OpenAI Guardrails
https://hiddenlayer.com/innovation-hub/same-model-different-hat
]-> tools to block/detect potentially harmful model behavior
// OpenAI’s Guardrails framework is a thoughtful attempt to provide developers with modular, customizable safety tooling, but its reliance on LLM-based judges highlights a core limitation of self-policing AI systems. True AI security demands independent validation layers, continuous red teaming, and visibility into how models interpret and act on inputs
#AIOps
#MLSecOps
Same Model, Different Hat:
Bypassing OpenAI Guardrails
https://hiddenlayer.com/innovation-hub/same-model-different-hat
]-> tools to block/detect potentially harmful model behavior
// OpenAI’s Guardrails framework is a thoughtful attempt to provide developers with modular, customizable safety tooling, but its reliance on LLM-based judges highlights a core limitation of self-policing AI systems. True AI security demands independent validation layers, continuous red teaming, and visibility into how models interpret and act on inputs
Forwarded from CyberSecurityTechnologies
RAG_Defender.pdf
629.9 KB
#tools
#RAG_Security
"Rescuing the Unpoisoned: Efficient Defense against Knowledge Corruption Attacks on RAG Systems", Nov. 2025.
]-> https://github.com/SecAI-Lab/RAGDefender
// RAGDefender - efficient defense mechanism designed to protect Retrieval-Augmented Generation (RAG) systems from knowledge corruption attacks such as PoisonedRAG, Blind, and GARAG. It detects and isolates poisoned documents in retrieved contexts without requiring additional model training or fine-tuning
#RAG_Security
"Rescuing the Unpoisoned: Efficient Defense against Knowledge Corruption Attacks on RAG Systems", Nov. 2025.
]-> https://github.com/SecAI-Lab/RAGDefender
// RAGDefender - efficient defense mechanism designed to protect Retrieval-Augmented Generation (RAG) systems from knowledge corruption attacks such as PoisonedRAG, Blind, and GARAG. It detects and isolates poisoned documents in retrieved contexts without requiring additional model training or fine-tuning
Forwarded from 御魂Hacker_🇨🇳-军火库 (黑 大帅🇨🇳)
https://github.com/itsOwen/CyberScraper-2077
A Powerful web scraper powered by LLM | OpenAI, Gemini & Ollama
#github #tools
A Powerful web scraper powered by LLM | OpenAI, Gemini & Ollama
#github #tools
❤1
Forwarded from CyberSecurityTechnologies
emb3d-stix-2.0.1.json
766.7 KB
#tools
#AIOps
#Red_Team_Tactics
"UDora: A Unified Red Teaming Framework against LLM Agents by Dynamically Hijacking Their Own Reasoning", Nov. 2025.
]-> https://github.com/AI-secure/UDora
// In this work, we present UDora, a unified red teaming framework designed for LLM agents that dynamically hijacks the agent's reasoning processes to compel malicious behavior
#AIOps
#Red_Team_Tactics
"UDora: A Unified Red Teaming Framework against LLM Agents by Dynamically Hijacking Their Own Reasoning", Nov. 2025.
]-> https://github.com/AI-secure/UDora
// In this work, we present UDora, a unified red teaming framework designed for LLM agents that dynamically hijacks the agent's reasoning processes to compel malicious behavior
👎1
Forwarded from CyberSecurityTechnologies
#tools
#AIOps
#Offensive_security
1⃣ NeuroSploitv2 - AI-Powered Pentesting Framework
// Leveraging the capabilities of LLMs, NeuroSploitv2 provides specialized agent roles that can analyze targets, identify vulnerabilities, plan exploitation strategies, and assist in defensive measures
2⃣ HexStrike AI MCP Agents v6.0
// Advanced AI-powered pentesting MCP framework with 150+ security tools and 12+ autonomous AI agents
3⃣ Offensive AI Agent Prompts
// Prompts for performing tests on your Kali Linux using Gemini-cli, ChatGPT, DeepSeek, CursorAI, Claude Code, and Copilot
#AIOps
#Offensive_security
1⃣ NeuroSploitv2 - AI-Powered Pentesting Framework
// Leveraging the capabilities of LLMs, NeuroSploitv2 provides specialized agent roles that can analyze targets, identify vulnerabilities, plan exploitation strategies, and assist in defensive measures
2⃣ HexStrike AI MCP Agents v6.0
// Advanced AI-powered pentesting MCP framework with 150+ security tools and 12+ autonomous AI agents
3⃣ Offensive AI Agent Prompts
// Prompts for performing tests on your Kali Linux using Gemini-cli, ChatGPT, DeepSeek, CursorAI, Claude Code, and Copilot
🔥1
Forwarded from Codeby
MASTER OSINT
Master OSINT — это комплексный, удобный для начинающих набор инструментов на Python, предназначенный для проведения расследований на основе открытых источников информации. Инструмент поддерживает множество методов расследования.
📕 Характиристика:
➖ Извлечение GPS-координат из EXIF-данных изображений.
➖ Исследование присутствия в социальных сетях на более чем 30 платформах по имени пользователя.
➖ Выявление утечек электронной почты и анализ общедоступных фрагментов текста.
➖ Проверка подлинности электронной почты с помощью Hunter.io, ReverseContact.com и Epieos.
➖ Извлечение метаданных из изображений/документов.
➖ Использование расширенных операторов поиска Google (Google Dorking).
➖ Доступ к архивным снимкам веб-сайтов через Wayback Machine.
➖ Определение IP-адреса и получение отчетов о нарушениях с помощью AbuseIPDB.
➖ Парсинг метаданных веб-сайтов и извлечение именованных сущностей с помощью spaCy NLP.
➖ Проверка и исследование телефонных номеров с информацией о операторе связи и типе телефона.
➖ Выполнение обратного поиска изображений с использованием популярных поисковых систем.
➖ Доступ к мощной геопространственной аналитике с помощью Google Satellite и OpenStreetMap.
💻 Установка:
📌 Запуск:
#tools #osint
➡️ Все наши каналы 💬 Все наши чаты ⚡️ Для связи с менеджером
Master OSINT — это комплексный, удобный для начинающих набор инструментов на Python, предназначенный для проведения расследований на основе открытых источников информации. Инструмент поддерживает множество методов расследования.
git clone https://gist.github.com/5fb49b007d48cf2717bc3c12958e47b5.git
mv 5fb49b007d48cf2717bc3c12958e47b5/ osintmaster/
cd osintmaster/
pip install waybackpy --break-system-packages
python3 Master_osint.py
#tools #osint
Please open Telegram to view this post
VIEW IN TELEGRAM
Forwarded from CyberSecurityTechnologies
#tools
#hardening
#MLSecOps
Detecting and Monitoring OpenClaw (clawdbot, moltbot)
1⃣ OpenClaw Detection Scripts
// Detection scripts for MDM deployment to identify OpenClaw installations on managed devices
2⃣ OpenClaw Telemetry Plugin
// Captures tool calls, LLM usage, agent lifecycle, and message events
3⃣ Advanced Cognitive Inoculation Prompt (ACIP)
// Fortifying LLMs against sophisticated prompt injection attacks
#hardening
#MLSecOps
Detecting and Monitoring OpenClaw (clawdbot, moltbot)
1⃣ OpenClaw Detection Scripts
// Detection scripts for MDM deployment to identify OpenClaw installations on managed devices
2⃣ OpenClaw Telemetry Plugin
// Captures tool calls, LLM usage, agent lifecycle, and message events
3⃣ Advanced Cognitive Inoculation Prompt (ACIP)
// Fortifying LLMs against sophisticated prompt injection attacks
Forwarded from CyberSecurityTechnologies
PIArena.pdf
1 MB
#tools
#MLSecOps
"PIArena: A Platform for Prompt Injection Evaluation",
Apr. 2026.
]-> Tool and datasets
#MLSecOps
"PIArena: A Platform for Prompt Injection Evaluation",
Apr. 2026.
]-> Tool and datasets
Forwarded from CyberSecurityTechnologies
#tools
#Blue_Team_Techniques
#Purple_Team_Exercises
AiSOC v.7.2.0
https://github.com/beenuar/AiSOC
// Open-source AI-powered Security Operations Center - alert fusion, purple-team drills, agent-assisted triage, MITRE ATT&CK investigation. MIT-licensed, self-hostable
#Blue_Team_Techniques
#Purple_Team_Exercises
AiSOC v.7.2.0
https://github.com/beenuar/AiSOC
// Open-source AI-powered Security Operations Center - alert fusion, purple-team drills, agent-assisted triage, MITRE ATT&CK investigation. MIT-licensed, self-hostable
Forwarded from CyberSecurityTechnologies
AgentWall.pdf
136.2 KB
#tools
#AIOps
"AgentWall: A Runtime Safety Layer for Local AI Agents", Mar. 2026.
]-> https://github.com/agentwall/Agentwall
// Run AI agents safely on your local machine
#AIOps
"AgentWall: A Runtime Safety Layer for Local AI Agents", Mar. 2026.
]-> https://github.com/agentwall/Agentwall
// Run AI agents safely on your local machine
❤1
Forwarded from CyberSecurityTechnologies
Remote_Agent_Takeover.pdf
5.1 MB
#tools
#AIOps
"Your WAF Blocked Us, That Was The Exploit:
Remote Agent Takeover via Cloudflare, Sentry, Datadog and Claude Zero-Day for data exfil and persistence",
DEF CON 34, 2026.
]-> Drop-in configs to harden Cursor and Claude Code against prompt injection
See also:
]-> DEF CON 34 - All Presentations
#AIOps
"Your WAF Blocked Us, That Was The Exploit:
Remote Agent Takeover via Cloudflare, Sentry, Datadog and Claude Zero-Day for data exfil and persistence",
DEF CON 34, 2026.
]-> Drop-in configs to harden Cursor and Claude Code against prompt injection
See also:
]-> DEF CON 34 - All Presentations