Search results for mcp security | Breaking Cybersecurity News | The Hacker News

Cybersecurity researchers have disclosed details of a new SmartLoader campaign that involves distributing a trojanized version of a Model Context Protocol ( MCP ) server associated with Oura Health to deliver an information stealer known as StealC . "The threat actors cloned a legitimate Oura MCP Server – a tool that connects AI assistants to Oura Ring health data – and built a deceptive infrastructure of fake forks and contributors to manufacture credibility," Straiker's AI Research (STAR) Labs team said in a report shared with The Hacker News. The end game is to leverage the trojanized version of the Oura MCP server to deliver the StealC infostealer, allowing the threat actors to steal credentials, browser passwords, and data from cryptocurrency wallets. SmartLoader, first highlighted by OALABS Research in early 2024, is a malware loader that's known to be distributed via fake GitHub repositories containing artificial intelligence (AI)-generated lures to giv...

Network firewalls are the workhorses of modern cybersecurity. They are trusted to protect the network, blocking malicious traffic and preventing intrusions and breaches. And for decades, network security teams have built controls around a relatively stable model: users connect to applications, applications exchange data, and security tools inspect packets, protocols, and destinations. Firewalls became exceptionally good at understanding where traffic was going and whether it should be allowed. But just as AI is reshaping every aspect of the business world, it's also had a monumental impact on how security teams view network traffic and the firewall. AI is driving new network dynamics that traditional security policies were never designed to govern. Employees, applications, and agents send prompts, call models, connect to services, and trigger actions across the environment, all activity most firewalls can't see, let alone understand. AI agents interact with external servic...

Cybersecurity researchers have discovered what has been described as the first-ever instance of a malicious Model Context Protocol ( MCP ) server spotted in the wild, raising software supply chain risks. According to Koi Security, a legitimate-looking developer managed to slip in rogue code within an npm package called " postmark-mcp " that copied an official Postmark Labs library of the same name. The malicious functionality was introduced in version 1.0.16, which was released on September 17, 2025. The actual "postmark-mcp" library, available on GitHub , exposes an MCP server to allow users to send emails, access and use email templates, and track campaigns using artificial intelligence (AI) assistants. The npm package in question has since been deleted from npm by the developer " phanpak ," who uploaded it to the repository on September 15, 2025, and maintains 31 other packages. The JavaScript library attracted a total of 1,643 downloads. ...

Cybersecurity researchers have discovered a critical "by design" weakness in the Model Context Protocol's ( MCP ) architecture that could pave the way for remote code execution and have a cascading effect on the artificial intelligence (AI) supply chain. "This flaw enables Arbitrary Command Execution (RCE) on any system running a vulnerable MCP implementation, granting attackers direct access to sensitive user data, internal databases, API keys, and chat histories," OX Security researchers Moshe Siman Tov Bustan, Mustafa Naamnih, Nir Zadok, and Roni Bar said in an analysis published last week. The cybersecurity company said the systemic vulnerability is baked into Anthropic's official MCP software development kit (SDK) across any supported language, including Python, TypeScript, Java, and Rust. In all, it affects more than 7,000 publicly accessible servers and software packages totaling more than 150 million downloads. At issue are unsafe defaults in...

Claude Code reads files, runs shell commands, invokes MCP tools, and acts through the credentials available on a developer’s machine. Anthropic’s new Compliance API endpoints give security teams their clearest view yet into that activity. They also expose a larger problem: activity logs alone cannot tell you whether an agent’s access is legitimate. AI has moved from the browser tab to the endpoint with harnesses like Claude Code. They run on developers' machines, execute bash commands locally, and connect to third parties via MCP servers, skills, and plugins. All this so the user can outsource labor to the machine and focus on designing, thinking, and creating. Local agents are not a niche category. They account for 68.6% of the AI agents Token Security discovers in customer environments, and they often inherit the employee's credentials, network position, and permissions. The shift to the endpoint has major implications for security. With Claude Code, there is no ce...

A set of three security vulnerabilities has been disclosed in mcp-server-git , the official Git Model Context Protocol ( MCP ) server maintained by Anthropic, that could be exploited to read or delete arbitrary files and execute code under certain conditions. "These flaws can be exploited through prompt injection, meaning an attacker who can influence what an AI assistant reads (a malicious README, a poisoned issue description, a compromised webpage) can weaponize these vulnerabilities without any direct access to the victim's system," Cyata researcher Yarden Porat said in a report shared with The Hacker News. Mcp-server-git is a Python package and an MCP server that provides a set of built-in tools to read, search, and manipulate Git repositories programmatically via large language models (LLMs). The security issues, which have been addressed in versions 2025.9.25 and 2025.12.18 following responsible disclosure in June 2025, are listed below - CVE-2025-68143 (CV...

Cybersecurity researchers have flagged a maximum-severity security flaw in Ruflo , an open-source agent meta-harness for Anthropic Claude Code and OpenAI Codex, that could result in unauthenticated remote code execution. The vulnerability, tracked as CVE-2026-59726 (CVSS score: 10.0), impacts all versions of the project before version 3.16.3. It has been codenamed RufRoot by Noma Security's research team, Noma Labs. Originally launched as Claude Flow, Ruflo is an AI multi-agent orchestration platform and harness that allows users to deploy multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. The project has more than 66,500 stars on GitHub. The crux of the vulnerability is that Ruflo exposed 233 tools, including shell command execution, database operations, agent management, and memory storage, through an unauthenticated Model Context Protocol (MCP) bridge that's open to the network by default. Specifically, the "docker-comp...

Microsoft on Tuesday unveiled the expansion of its Sentinel Security Incidents and Event Management solution (SIEM) as a unified agentic platform with the general availability of the Sentinel data lake. In addition, the tech giant said it's also releasing a public preview of Sentinel Graph and Sentinel Model Context Protocol ( MCP ) server to turn telemetry into a security graph and allow AI agents access an organization's security context in a standardized manner. "With graph-based context, semantic access, and agentic orchestration, Sentinel gives defenders a single platform to ingest signals, correlate across domains, and empower AI agents built in Security Copilot, VS Code using GitHub Copilot, or other developer platforms," Vasu Jakkal, corporate vice president at Microsoft Security, said in a post shared with The Hacker News. Microsoft released Sentinel data lake in public preview earlier this July as a purpose-built, cloud-native tool to ingest, manage...

A lot of security problems still begin with someone doing a completely normal thing. Cloning a repo. Answering a call. Leaving a box exposed. Trusting the default. That pretty much covers the mood this week. Old bugs are back, supply chains are getting stranger, and some exploit paths are so short you wonder what was supposed to stop them in the first place. That’s only part of it. Here’s everything else that made the Monday recap. ⚡ Threat of the Week Anthropic's Model Attempts to Poison Open-Source Project — A new evaluation conducted by the U.K. AI Security Institute (AISI) found that AI models with access to the internet reached out into the real world to target individuals and organizations autonomously across 10 of the total of 122 runs. Of 19 such actions recorded, 17 originated from Anthropic's Mythos 5 and the remaining two involved OpenAI's GPT-5.6-Sol with cyber classifiers. In the most serious case, Anthropic's Claude Mythos 5 spent 34 hours trying...

New research from CrowdStrike has revealed that DeepSeek's artificial intelligence (AI) reasoning model DeepSeek-R1 produces more security vulnerabilities in response to prompts that contain topics deemed politically sensitive by China. "We found that when DeepSeek-R1 receives prompts containing topics the Chinese Communist Party (CCP) likely considers politically sensitive, the likelihood of it producing code with severe security vulnerabilities increases by up to 50%," the cybersecurity company said . The Chinese AI company previously attracted national security concerns, leading to a ban in many countries. Its open-source DeepSeek-R1 model was also found to censor topics considered sensitive by the Chinese government, refusing to answer questions about the Great Firewall of China or the political status of Taiwan, among others. In a statement released earlier this month, Taiwan's National Security Bureau warned citizens to be vigilant when using Chinese-m...

A malicious MCP server could trick an application built on the official  MCP Python SDK  into handing over the OAuth credentials it uses to log in to a real service, the SDK's maintainers said in a security advisory. Affected versions sent the client secret, the authorization code, and the PKCE proof key to a token endpoint the attacker controlled. The fix is in versions 1.30.0 and 2.2.0. The Model Context Protocol (MCP) is an open standard for connecting AI applications to outside tools and data, and this package is its official Python SDK for building MCP servers and clients. With the stolen credentials, the attacker can request a valid access token from the real login service. Cycode, the security firm that reported the flaw , demonstrated that full exchange in a test and says the resulting token carries whatever permissions the app was granted. The client secret is long-lived, so it keeps working until it is changed. The flaw is rated high (7.5) for the two prov...

On January 31, 2026, researchers disclosed that Moltbook, a social network built for AI agents, had left its database wide open, exposing 35,000 email addresses and 1.5 million agent API tokens across 770,000 active agents. The more worrying part sat inside the private messages. Some of those conversations held plaintext third-party credentials, including OpenAI API keys shared between agents, stored in the same unencrypted table as the tokens needed to hijack the agent itself. This is the shape of a toxic combination: a permission breakdown between two or more applications, bridged by an AI agent, integration, or OAuth grant, that no single application owner ever authorized as its own risk surface. Moltbook's agents sat at that bridge, carrying credentials for their host platform and for the outside services their users had wired them into, in a place that neither platform owner had line of sight into. Most SaaS access reviews still examine one application at a time, which is...

Over 30 security vulnerabilities have been disclosed in various artificial intelligence (AI)-powered Integrated Development Environments (IDEs) that combine prompt injection primitives with legitimate features to achieve data exfiltration and remote code execution. The security shortcomings have been collectively named IDEsaster by security researcher Ari Marzouk (MaccariTA), who discovered them over the last six months. They affect popular IDEs and extensions such as Cursor, Windsurf, Kiro.dev, GitHub Copilot, Zed.dev, Roo Code, Junie, and Cline, among others. Of these, 24 have been assigned CVE identifiers. "I think the fact that multiple universal attack chains affected each and every AI IDE tested is the most surprising finding of this research," Marzouk told The Hacker News. "All AI IDEs (and coding assistants that integrate with them) effectively ignore the base software (IDE) in their threat model. They treat their features as inherently safe because they’ve...

A security weakness has been disclosed in the artificial intelligence (AI)-powered code editor Cursor that could trigger code execution when a maliciously crafted repository is opened using the program. The issue stems from the fact that an out-of-the-box security setting is disabled by default, opening the door for attackers to run arbitrary code on users' computers with their privileges. "Cursor ships with Workspace Trust disabled by default, so VS Code-style tasks configured with runOptions.runOn: 'folderOpen' auto-execute the moment a developer browses a project," Oasis Security said in an analysis. "A malicious .vscode/tasks.json turns a casual 'open folder' into silent code execution in the user's context." Cursor is an AI-powered fork of Visual Studio Code, which supports a feature called Workspace Trust to allow developers to safely browse and edit code regardless of where it came from or who wrote it. With this option disab...

AI coding agents are changing how quickly developers can build and ship software as well as how quickly credentials can become exposed. According to GitGuardian’s 2026 State of Secrets Sprawl Report , commits identified as AI-assisted are leaking secrets at approximately twice the rate of human-written ones. Most of the fastest-growing categories of leaked credentials are now connected to AI services, meaning the tools meant to advance development are also accelerating the exposure of the keys development relies on. This isn’t a new vulnerability; what is new is AI changing the scale and pace at which those mistakes can happen. A coding agent can read an entire project, modify files, generate configurations and interact with external services in the time a developer might take to review a single pull request. The main issue isn’t that AI agents sometimes encounter secrets but that many of those secrets were never designed for an environment in which software can act autonomously. A...

Cybersecurity researchers have disclosed details of a vulnerability in OpenAI ChatGPT that leverages the artificial intelligence (AI) assistant's implicit trust in Markdown links and images to trigger prompt injections and open the door to phishing attacks. The technique has been codenamed ChatGPhish by Permiso Security. "The chatgpt.com response renderer trusts Markdown links and Markdown image URLs that originated from a third-party page the assistant has just summarized. It auto-fetches those images and surfaces those links as live, clickable elements inside the trusted assistant UI," security researcher Andi Ahmeti said in a report shared with The Hacker News. In a hypothetical attack scenario, a bad actor can append a small payload to any web page that the victim later prompts ChatGPT to summarize, causing it to leak their IP, User-Agent, and Referer details when attacker-hosted images embedded in the page are automatically fetched when the answer is rendered...

Criminals don’t need to be clever all the time; they just follow the easiest path in: trick users, exploit stale components, or abuse trusted systems like OAuth and package registries. If your stack or habits make any of those easy, you’re already a target. This week’s ThreatsDay highlights show exactly how those weak points are being exploited — from overlooked misconfigurations to sophisticated new attack chains that turn ordinary tools into powerful entry points. Lumma Stealer Stumbles After Doxxing Drama Decline in Lumma Stealer Activity After Doxxing Campaign The activity of the Lumma Stealer (aka Water Kurita) information stealer has witnessed a "sudden drop" since last months after the identities of five alleged core group members were exposed as part of what's said to be an aggressive underground exposure campaign dubbed Lumma Rats since late August 2025. The targeted individuals are affiliated with the malware's development and administ...

As 2025 draws to a close, security professionals face a sobering realization: the traditional playbook for web security has become dangerously obsolete. AI-powered attacks, evolving injection techniques, and supply chain compromises affecting hundreds of thousands of websites forced a fundamental rethink of defensive strategies. Here are the five threats that reshaped web security this year, and why the lessons learned will define digital protection for years to come. 1. Vibe Coding Natural language coding, " vibe coding " , transformed from novelty to production reality in 2025, with nearly 25% of Y Combinator startups using AI to build core codebases. One developer launched a multiplayer flight simulator in under three hours, eventually scaling it to 89,000 players and generating thousands in monthly revenue. The Result Code that functions perfectly yet contains exploitable flaws, bypassing traditional security tools. AI generates what you ask for, not what you forget...

Microsoft on Tuesday released patches for a set of 84 new security vulnerabilities affecting various software components, including two that have been listed as publicly known. Of these, eight are rated Critical, and 76 are rated Important in severity. Forty-six of the patched vulnerabilities relate to privilege escalation, followed by 18 remote code execution, 10 information disclosure, four spoofing, four denial-of-service, and two security feature bypass flaws. The fixes are in addition to 10 vulnerabilities that have been addressed in its Chromium-based Edge browser since the release of the February 2026 Patch Tuesday update . The two publicly disclosed zero-days are CVE-2026-26127 (CVSS score: 7.5), a denial-of-service vulnerability in .NET, and CVE-2026-21262 (CVSS score: 8.8), an elevation of privilege vulnerability in SQL Server. The vulnerability with the highest CVSS score in this month's update is a critical remote code execution flaw in the Microsoft Devices ...

ThreatsDay Bulletin is back on The Hacker News, and this week feels off in a familiar way. Nothing loud, nothing breaking everything at once. Just a lot of small things that shouldn’t work anymore but still do. Some of it looks simple, almost sloppy, until you see how well it lands. Other bits feel a little too practical, like they’re already closer to real-world use than anyone wants to admit. And the background noise is getting louder again, the kind people usually ignore. A few stories are clever in a bad way. Others are just frustratingly avoidable. Overall, it feels like quiet pressure is building in places that matter. Skim it or read it properly, but don’t skip this one. Emerging RaaS exploiting FortiGate flaws The Gentlemen RaaS Detailed Group-IB has shed light on the various tactics adopted by The Gentlemen, a nascent Ransomware-as-a-Service (RaaS) operation that consists of about 20 members. It originated f...