Search results for New agent | Breaking Cybersecurity News | The Hacker News

The Iranian threat actor tracked as APT34 has been linked to a new phishing attack that leads to the deployment of a variant of a backdoor called  SideTwist . "APT34 has a high level of attack technology, can design different intrusion methods for different types of targets, and has supply chain attack capability," NSFOCUS Security Labs  said  in a report published last week. APT34, also known by the names Cobalt Gypsy, Hazel Sandstorm (formerly Europium), Helix Kitten, and OilRig, has a  track record  of targeting telecommunications, government, defense, oil and financial services verticals in the Middle East since at least 2014 via spear-phishing lures that culminate in the deployment of various backdoors. One of the key traits of the hacking outfit is its ability to create new and updated tools to minimize the odds of detection and gain a foothold on compromised hosts for extended periods of time. SideTwist was  first documented  as used by AP...

Security researchers on Tuesday uncovered new delivery and evasion techniques adopted by Agent Tesla remote access trojan (RAT) to get around defense barriers and monitor its victims. Typically spread through social engineering lures, the Windows spyware not only now targets Microsoft's Antimalware Scan Interface ( AMSI ) in an attempt to defeat endpoint protection software, it also employs a multi-stage installation process and makes use of Tor and Telegram messaging API to communicate with a command-and-control (C2) server. Cybersecurity firm Sophos , which observed two versions of Agent Tesla — version 2 and version 3 — currently in the wild, said the changes are yet another sign of Agent Tesla's constant evolution designed to make a sandbox and static analysis more difficult. "The differences we see between v2 and v3 of Agent Tesla appear to be focused on improving the success rate of the malware against sandbox defenses and malware scanners, and on providing more...

Cybersecurity researchers have disclosed a zero-click flaw in OpenAI ChatGPT's Deep Research agent that could allow an attacker to leak sensitive Gmail inbox data with a single crafted email without any user action. The new class of attack has been codenamed ShadowLeak by Radware. Following responsible disclosure on June 18, 2025, the issue was addressed by OpenAI in early August. "The attack utilizes an indirect prompt injection that can be hidden in email HTML (tiny fonts, white-on-white text, layout tricks) so the user never notices the commands, but the agent still reads and obeys them," security researchers Zvika Babo, Gabi Nakibly, and Maor Uziel said . "Unlike prior research that relied on client-side image rendering to trigger the leak, this attack leaks data directly from OpenAI's cloud infrastructure, making it invisible to local or enterprise defenses." Launched by OpenAI in February 2025, Deep Research is an agentic capability built into ...

For years, the two most popular methods for internal scanning: agent-based and network-based were considered to be about equal in value, each bringing its own strengths to bear. However, with remote working now the norm in most if not all workplaces, it feels a lot more like agent-based scanning is a must, while network-based scanning is an optional extra. This article will go in-depth on the strengths and weaknesses of each approach, but let's wind it back a second for those who aren't sure why they should even do internal scanning in the first place. Why should you perform internal vulnerability scanning? While  external vulnerability scanning  can give a great overview of what you look like to a hacker, the information that can be gleaned without access to your systems can be limited. Some serious vulnerabilities can be discovered at this stage, so it's a must for many organizations, but that's not where hackers stop.  Techniques like phishing, targeted malware,...

Cybersecurity researchers have discovered a new post-exploitation technique in Amazon Web Services (AWS) that allows the AWS Systems Manager Agent (SSM Agent) to be run as a remote access trojan on Windows and Linux environments "The SSM agent, a legitimate tool used by admins to manage their instances, can be re-purposed by an attacker who has achieved high privilege access on an endpoint with SSM agent installed, to carry out malicious activities on an ongoing basis," Mitiga researchers Ariel Szarf and Or Aspir  said  in a report shared with The Hacker News. "This allows an attacker who has compromised a machine, hosted on AWS or anywhere else, to maintain access to it and perform various malicious activities." SSM Agent is a  software  installed on Amazon Elastic Compute Cloud (Amazon EC2) instances, on-premise servers, and virtual machines, making it possible for administrators to update, manage, and configure their AWS resources through a unified interface....

Cybersecurity researchers have disclosed a now-patched security flaw in LangChain's LangSmith platform that could be exploited to capture sensitive data, including API keys and user prompts. The vulnerability, which carries a CVSS score of 8.8 out of a maximum of 10.0, has been codenamed AgentSmith by Noma Security. LangSmith is an observability and evaluation platform that allows users to develop, test, and monitor large language model (LLM) applications, including those built using LangChain. The service also offers what's called a LangChain Hub , which acts as a repository for all publicly listed prompts, agents, and models. "This newly identified vulnerability exploited unsuspecting users who adopt an agent containing a pre-configured malicious proxy server uploaded to 'Prompt Hub,'" researchers Sasi Levi and Gal Moyal said in a report shared with The Hacker News. "Once adopted, the malicious proxy discreetly intercepted all user communicatio...

AI agents are rapidly becoming a core part of the enterprise, being embedded across enterprise workflows, operating with autonomy, and making decisions about which systems to access and how to use them. But as agents grow in power and autonomy, so do the risks and threats.  Recent studies show 80% of companies have already experienced unintended AI agent actions, from unauthorized system access to data leaks. These incidents aren't edge cases. They are the inevitable outcome of deploying AI agents at scale without purpose-built security mechanisms. Traditional IAM wasn't designed for this. Agents move too fast, operate 24/7, while relying on non-human identities (NHIs) to define precisely what they can and can't do. How can organizations possibly secure what they cannot see or control? To address this challenge, a new approach is needed—one that enables secure-by-design AI agent deployment across the enterprise. Enter: Astrix's Agent Control Plane (ACP) Astrix's AI Agent Cont...

A new variant of the  Agent Tesla  malware has been observed delivered via a lure file with the  ZPAQ compression format  to harvest data from several email clients and nearly 40 web browsers. "ZPAQ is a file compression format that offers a better compression ratio and journaling function compared to widely used formats like ZIP and RAR," G Data malware analyst Anna Lvova  said  in a Monday analysis. "That means that ZPAQ archives can be smaller, saving storage space and bandwidth when transferring files. However, ZPAQ has the biggest disadvantage: limited software support." First appearing in 2014, Agent Tesla is a  keylogger  and  remote access trojan  (RAT) written in .NET that's  offered  to other threat actors as part of a malware-as-a-service (MaaS) model. It's often used as a first-stage payload, providing remote access to a compromised system and utilized to download more sophisticated second-stage tools such as ...

Cybersecurity researchers have disclosed a now-patched, high-severity security flaw in Cursor, a popular artificial intelligence (AI) code editor, that could result in remote code execution (RCE). The vulnerability, tracked as CVE-2025-54135 (CVSS score: 8.6), has been addressed in version 1.3 released on July 29, 2025. It has been codenamed CurXecute by Aim Labs, which previously disclosed EchoLeak . "Cursor runs with developer‑level privileges, and when paired with an MCP server that fetches untrusted external data, that data can redirect the agent's control flow and exploit those privileges," the Aim Labs Team said in a report shared with The Hacker News. "By feeding poisoned data to the agent via MCP, an attacker can gain full remote code execution under the user privileges, and achieve any number of things, including opportunities for ransomware, data theft, AI manipulation and hallucinations, etc." In other words, the remote code execution can trigg...

Two malicious packages discovered in the npm package repository have been found to conceal an open source information stealer malware called  TurkoRat . The packages – named nodejs-encrypt-agent and nodejs-cookie-proxy-agent – were collectively downloaded approximately 1,200 times and were available for more than two months before they were identified and taken down. ReversingLabs, which broke down the details of the campaign, described TurkoRat as an information stealer capable of harvesting sensitive information such as login credentials, website cookies, and data from cryptocurrency wallets.  While nodejs-encrypt-agent came fitted with the malware inside, nodejs-cookie-proxy-agent was found to disguise the trojan as a dependency under the name axios-proxy. nodejs-encrypt-agent was also engineered to masquerade as another legitimate npm module known as  agent-base , which has been downloaded over 25 million times to date. The list of the rogue packages and their a...

AI agents promise to automate everything from financial reconciliations to incident response. Yet every time an AI agent spins up a workflow, it has to authenticate somewhere; often with a high-privilege API key, OAuth token, or service account that defenders can't easily see. These "invisible" non-human identities (NHIs) now outnumber human accounts in most cloud environments, and they have become one of the ripest targets for attackers. Astrix's Field CTO Jonathan Sander put it bluntly in a recent Hacker News webinar : "One dangerous habit we've had for a long time is trusting application logic to act as the guardrails. That doesn't work when your AI agent is powered by LLMs that don't stop and think when they're about to do something wrong. They just do it." Why AI Agents Redefine Identity Risk Autonomy changes everything: An AI agent can chain multiple API calls and modify data without a human in the loop. If the underlying credential is exposed or overprivileged, each addit...

ChatGPT and similar large language models (LLMs) have added further complexity to the ever-growing online threat landscape. Cybercriminals no longer need advanced coding skills to execute fraud and other damaging attacks against online businesses and customers, thanks to bots-as-a-service, residential proxies, CAPTCHA farms, and other easily accessible tools.  Now, the latest technology damaging businesses' bottom line is  ChatGPT . Not only have ChatGPT, OpenAI, and other LLMs raised ethical issues by  training their models  on scraped data from across the internet. LLMs are negatively impacting enterprises' web traffic, which can be extremely damaging to business.  3 Risks Presented by LLMs, ChatGPT, & ChatGPT Plugins Among the threats ChatGPT and ChatGPT plugins can pose against online businesses, there are three key risks we will focus on: Content theft  (or republishing data without permission from the original source)can hurt the authority,...