Tag: open-source
-
Fake Zoom Installer Uses .NET Downloader to Deploy Overlord RAT on macOS
A cross-platform malware campaign that disguises itself as a legitimate Zoom installer to deploy Overlord, an open-source remote access trojan (RAT), on both macOS and Windows machines. Documented by Jamf, unlike most macOS malware, which typically relies on Go or Rust for cross-platform reach, this campaign’s first-stage downloader, a macOS ARM64 Mach-O binary named ZoomMeetings,…
-
Black Hat 2026: Open-source tool makes red teaming AI agents up to 125x cheaper
First seen on scworld.com Jump to article: www.scworld.com/news/black-hat-2026-open-source-tool-makes-red-teaming-ai-agents-up-to-125x-cheaper
-
1Password Research Finds AI-Generated Vulnerability Patches Often Leave Bugs Behind
1Password’s Off-by-1 Labs has released research showing that large language models frequently produce vulnerability patches that look plausible but fail to fix the underlying flaw. The study, titled Frontier Models’ Vulnerability Patches are Often F.L.A.W.E.D., evaluated two cyber-capable reasoning models against six recently disclosed vulnerabilities in open-source software. Researchers generated 6,080 patches across the test..…
-
Agentic AI for Cyber Defenders: What Security Teams Built at Black Hat USA 2026
Tags: ai, automation, conference, control, credentials, cve, cyber, cybersecurity, data, data-breach, defense, detection, exploit, flaw, group, iam, intelligence, ISO-27001, mitigation, network, nvidia, offense, open-source, RedTeam, risk, skills, soc, technology, threat, tool, usa, vulnerabilityAgentic AI armed attackers first, but it also put real building power in defenders’ hands. Here’s what security practitioners built in two days at Black Hat USA 2026, and how the CyberAgents Exchange keeps that work compounding long after the event. Key takeaways Building defensive cybersecurity tooling no longer requires a developer. Agentic tooling drove…
-
Growing Up The Hard Way
Tags: open-sourceOpen Source had a great childhood.For two decades it got to be a kid. It ran around barefoot, gave everything away, trusted strangers, and never once thought about who was watching. It ran the kind of lemonade stand that took IOUs from anyone who wandered up, take what you need, pay me back whenever, no…
-
Critical Paperclip AI Agent Flaws Allow Unauthenticated Remote Code Execution
Critical vulnerabilities in the open-source Paperclip AI-agent orchestration platform could allow attackers to execute commands remotely on exposed servers or on a developer’s local machine. These flaws arise from broken authorization boundaries across agent imports and API routes, as well as trust assumptions for localhost. Paperclip is designed to coordinate autonomous agents across >>companies,<< with…
-
Cloudflare Launches Open-Source OS to Secure AI Agents’ Access to Internal Data
Cloudflare has open-sourced Cloudflare OS, a platform designed to provide enterprise AI agents with controlled access to internal systems, company context, and workflows without exposing long-lived credentials or bypassing access controls. This release addresses a significant security challenge for enterprises: while agents need access to business data and tools to be effective, conventional API keys…
-
OSINT collection techniques using legitimate tools
Open source intelligence, usually shortened to OSINT, is the practice of collecting and analysing information that is already publicly available. In a defensive context, that can include company websites, social media posts, public registers, DNS records, certificate logs, archived web pages, document metadata, and other sources that are accessible without bypassing controls or impersonating anyone….…
-
Sicherheitstest mit Mythos 5 – KI-Agent wollte Malware in Open-Source-Projekt einschleusen
Bei einem Sicherheitstest versuchte ein auf Mythos 5 basierender Agent, Malware in ein echtes Open-Source-Projekt einzuschleusen. First seen on computerbase.de Jump to article: www.computerbase.de/news/apps/sicherheitstest-mit-mythos-5-ki-agent-wollte-malware-in-open-source-projekt-einschleusen.98721
-
Cloudflare OS goes open source with a record of everything its agents read
Tags: open-sourceCloudflare open sourced Cloudflare OS, the agent platform whose first version its own employees have used since May. Every resource an agent reads gets recorded, the record … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/08/06/cloudflare-os-open-source/
-
BSidesSF 2026 Building An Open Source Security Project With 1M+ Installations
Tags: open-sourcePresenters: Fletcher Heisler, Marcelo Elizeche Landó Our thanks to Security BSides San Francisco for publishing their Creators, Authors and Presenter’s outstanding BSidesSF 2026 content on the Organizations’ YouTube Channel. Permalink First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/08/bsidessf-2026-building-an-open-source-security-project-with-1m-installations/
-
Mini Shai-Hulud npm Attack: More Than 2,200 Components Impacted
Tags: access, ai, attack, breach, cloud, container, control, credentials, data, data-breach, github, guide, infection, intelligence, kubernetes, malicious, malware, microsoft, open-source, risk, sbom, service, software, threat, update<div cla TL;DR A new wave of the Shai-Hulud malicious package campaign emerged on npm, with 2,225 software component versions impacted. The malware executes through a malicious preinstall hook, steals npm, GitHub, cloud, Kubernetes, Vault, CI/CD, and other credentials, then uses stolen publishing access to compromise additional packages. Organizations that installed an affected version should…
-
XM Cyber Releases Open-Source Tools for Hunting macOS and Oracle Cloud Exposures
XM Cyber has announced new open-source exposure-hunting tools for macOS endpoints and Oracle Cloud Infrastructure. The release, issued from Black Hat USA and DEF CON, says the tools are intended to help security teams uncover and validate complex attack paths. The macOS tool examines weaknesses that can allow an attacker to move from an initial..…
-
Paperclip AI Flaws Let Attackers Run Host Commands via Malicious Agent Imports
Two security flaws in Paperclip could let attackers execute commands on a network server or a developer’s computer. Paperclip is an open-source control plane for teams of artificial intelligence (AI) agents, and both paths rely on importing a malicious agent and starting it.A third flaw could expose sensitive data and control-plane details through application programming…
-
What Gold Eagle Validates, and What Your Organization Still Has to Own (July 2026)
Tags: ai, business, compliance, data, finance, framework, government, intelligence, open-source, update, vulnerabilityWhat Gold Eagle Validates, and What Your Organization Still Has to Own (July 2026) The federal government just stood up a clearinghouse to find and validate vulnerabilities faster, with AI doing the finding. That is not the same thing as a clearinghouse that owns the consequence when a patch does not land in time. Key…
-
Open-source software’s archenemy TeamPCP goes back further than anyone thought
Oligo Security uncovered evidence of a long operational history, including multiple previous attacks it traced to the same attacker infrastructure and tools. First seen on cyberscoop.com Jump to article: cyberscoop.com/teampcp-long-active-history-2020-oligo-security/
-
Code review used to be the only way to catch these bugs
An automated system called NOVA read the source code of 3,915 open-source projects over two months and came back with 14,090 vulnerabilities, each one confirmed through the … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/08/05/code-review-ai-vulnerability-discovery/
-
Claude Mythos 5 Tried to Backdoor a Real Open-Source Project in Testing, Then Vouched for Itself
An agent running Anthropic’s Claude Mythos 5 spent 34 hours trying to get a malware dropper merged into a real open-source project during a cyber evaluation by the UK’s AI Security Institute.When a bystander publicly warned that the code was malicious, the agent denied it, force-pushed a rewritten branch history to erase the evidence, and…
-
OpenAI und Anthropic: KI-Modelle hacken sich erneut durchs Internet
KI-Modelle von OpenAI und Anthropic haben bei Tests echte Ziele attackiert. Sie wollten unter anderem Schadcode in ein Open-Source-Projekt einschleusen. First seen on golem.de Jump to article: www.golem.de/news/openai-und-anthropic-ki-modelle-hacken-sich-erneut-durchs-internet-2608-211608.html
-
Google Warns Open-Source Attacks Will Reach New Heights
Google Says Open-Source Compromises Are Easier to Scale and Replicate. Compromising the open-source supply chain is easy to do and spreads more quickly than traditional supply-chain attacks, making it a lucrative tactic that will continue to expand, warned Google. One of the largest open-source supply-chain attacks involved a North Korean threat actor. First seen on…
-
CISA lays out new guidance for using open-source software
The US Cybersecurity and Infrastructure Security Agency (CISA) has published the Open Source Software: Security Principles and Practices guide, which provides federal agencies … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/08/03/cisa-oss-security-guidance/
-
Shadow AI und autonome Agenten: Die unterschätzte Gefahr aus öffentlichen KI-Repositorien
Open-Source-Modelle und KI-Agenten schaffen neue Lieferkettenrisiken. Unternehmen brauchen Governance, Transparenz und menschliche Kontrolle. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/shadow-ai-und-autonome-agenten-die-unterschaetzte-gefahr-aus-oeffentlichen-ki-repositorien/a45980/
-
SkillSpector: NVIDIA’s open-source security scanner for AI agent skills
SkillSpector is an open-source scanner from NVIDIA that reads an agent skill and tells you whether to install it. Point it at a directory, a zip file, a single SKILL.md, or a … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/08/03/skillspector-open-source-agent-skill-security-scanner/
-
Week in review: Claude breached three companies during tests, AD CS domain-takeover PoC released
Here’s an overview of some of last week’s most interesting news, articles, interviews and videos: Nono: Open-source sandbox for AI agents AI coding agents run with the same … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/08/02/week-in-review-claude-breached-three-companies-during-tests-ad-cs-domain-takeover-poc-released/
-
To Ban or Not Ban Chinese Open-Weight AI Models
Tags: ai, backdoor, china, control, cybersecurity, data, defense, finance, government, infrastructure, international, malicious, microsoft, military, network, nvidia, open-source, openai, regulation, risk, software, supply-chain, technology, usaShould the US ban American companies from using Chinese open-weight AI models? That is the ugly question. US officials have openly expressed concerns and a desire to implement regulations. The technology community has aggressively responded, with over 20 leading AI companies, including Microsoft, Nvidia, Meta, and Dell, urging legislators not to rush imposing restrictions on…
-
Hacker uses DeepSeek AI to autonomously attack vulnerable servers
A Chinese-speaking threat actor is using the DeepSeek AI model and the open-source Hermes Agent to conduct autonomous cyberattacks on exposed servers with limited human involvement. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/security/hacker-uses-deepseek-ai-to-autonomously-attack-vulnerable-servers/
-
Amazon pins multiple open source compromises on North Korea
Amazon threat researchers found one threat actor behind four distinct open source compromises, including the March 2026 incident affecting the axios NPM library First seen on computerweekly.com Jump to article: www.computerweekly.com/news/366646561/Amazon-pins-multiple-open-source-compromises-on-North-Korea
-
Chinese Hacker Commands DeepSeek via Telegram to Launch Autonomous Attacks
Palo Alto Networks’ Unit 42 says a Chinese-speaking threat actor used DeepSeek through the open-source Hermes Agent framework to launch attacks autonomously.After an initial Telegram instruction, the agent found internet-facing systems and selected public exploits. The researchers recovered no further operator input in the session.The operator, tracked through the aliases knaithe and KnYuan, First seen…
-
Defining Community Open Source Is Harder Than It Looks
<div cla Exemptions sound relatively simple until you try to make them fair. First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/07/defining-community-open-source-is-harder-than-it-looks/