NadMesh: A New Threat in the AI Service Landscape

In early July, security researchers spotted a new Go-based botnet named NadMesh. It’s not your run-of-the-mill malware. This one has a singular focus: hunting exposed AI services to pilfer cloud keys and Kubernetes tokens. The operator’s own dashboard reportedly shows 3,811 unique AWS keys already harvested.

That’s a staggering number. And it underscores a growing problem — teams stand up AI tools fast, often skipping security configurations. NadMesh exploits that haste.

How NadMesh Finds Its Targets

The botnet uses a Shodan harvester to keep its scan queue stocked. It’s not scanning randomly. It targets specific platforms: ComfyUI, Ollama, n8n, Open WebUI, Langflow, and Gradio. These are the image generators, local model runners, and workflow builders that developers love for their speed and ease.

But speed comes at a cost. Many of these services are deployed without authentication or proper network segmentation. That makes them low-hanging fruit for attackers.

The Role of Shodan in NadMesh’s Operations

Shodan is a search engine for internet-connected devices. NadMesh uses it to identify vulnerable AI services exposed to the internet. Once found, the botnet moves in, exploiting known weaknesses or misconfigurations.

This isn’t a sophisticated zero-day exploit. It’s basic hygiene failure — services left open, default credentials, unpatched versions. NadMesh simply capitalizes on that.

What NadMesh Steals and Why It Matters

The primary targets are AWS keys and Kubernetes tokens. These credentials grant access to cloud infrastructure. With them, an attacker can spin up resources, exfiltrate data, or launch further attacks.

The 3,811 AWS keys on the operator’s dashboard represent real organizations. Each key is a potential entry point. And Kubernetes tokens? They can unlock entire clusters, exposing sensitive workloads and secrets.

This isn’t just about data theft. It’s about control. Once inside, attackers can use these resources for cryptomining, ransomware, or as a launchpad for bigger operations.

Protecting Your AI Services from NadMesh

So, what can you do? Start by auditing your exposed services. If you’re running ComfyUI, Ollama, or any of the targeted platforms, check if they’re accessible from the internet. They shouldn’t be.

• Restrict access with firewalls or VPNs.
• Enable authentication, even for internal tools.
• Regularly update to patch known vulnerabilities.
• Monitor network traffic for unusual activity.

These steps sound basic, but they’re often overlooked. NadMesh thrives on that oversight.

Securing Cloud Keys and Kubernetes Tokens

Beyond service exposure, pay attention to credential management. Rotate AWS keys regularly. Use short-lived tokens for Kubernetes. Implement least-privilege access. If a key is compromised, limit the damage.

Consider using secrets management tools. They centralize control and add an extra layer of security. It’s not foolproof, but it raises the bar.

The Bigger Picture: AI Security Is a Moving Target

NadMesh is a symptom of a larger issue. AI services are proliferating faster than security practices can keep up. Teams deploy tools for productivity, but they forget the basics.

The result? A botnet like NadMesh finds thousands of exposed services and walks away with thousands of cloud keys. It’s a wake-up call.

For more on securing your infrastructure, check out our guide on protecting cloud environments from botnet attacks. And if you’re using Kubernetes, you might want to read about Kubernetes security best practices.

Also, stay informed on AI service vulnerabilities to keep your deployments safe.

Don’t wait for the next NadMesh. Secure your AI services today.

Milan Startup’s Bet: Assume the Breach, Then Trap the Attacker

Most security tools try to keep hackers out. Beelzebub, an Italian AI-native cybersecurity startup, operates on a darker assumption: the bad guys are already in. And instead of just detecting them, the company wants to trap them.

That bet just got a serious cash infusion. The Milan-based firm announced it raised €3 million (roughly $3.4 million) in a seed round led by VC United Ventures. With this latest injection, the company’s total funding now stands at $3.8 million.

The Beelzebub Platform: A Closed Loop for Intruders

Founded in 2025, Beelzebub has built a platform that fuses red-team and blue-team tactics into a single response system. The core idea is simple: if you can’t stop every intrusion, make the intrusion itself a losing game for the attacker.

The platform’s approach is built on a few key pillars:

• Continuous adversary emulation – It actively probes for attack paths before real criminals exploit them.
• Runtime deception technology – LLM-powered traps that lure attackers into a controlled environment.
• Autonomous threat intelligence – When an attack is intercepted, the system automatically converts it into actionable defenses.
• AI analyst – Dissects malware, generates full incident reports, and contains affected systems instantly.

The whole loop is designed to isolate a threat, trigger incident response, and learn from the encounter — all without waiting for a human analyst to catch up.

On-Premises for the Paranoid (and the Regulated)

One of the more interesting angles is deployment flexibility. The platform works as a SaaS, but it can also run entirely on-premises. That’s a critical feature for organizations in sensitive environments — think critical infrastructure or government-adjacent sectors — that can’t send data to cloud-based tools.

NIS2-Ready and Backed by a Global Threat Intel Network

Beelzebub isn’t just building traps; it’s also plugged into a live intelligence feed from over 60 independent researchers worldwide. That’s a significant network for a startup that’s barely a year old.

The platform is also designed to be NIS2-ready out of the box. For European organizations scrambling to meet the EU’s updated cybersecurity directive, that’s a selling point that could open doors.

Where the Money Goes: Rome, San Francisco, and a Bigger Research Team

Founder and CEO Mario Candela has clear plans for the fresh capital. The company will expand its research team, open new offices in Rome and San Francisco, and aggressively pursue clients across Europe — with a focus on NIS2-regulated organizations.

“Cybersecurity is a nonstop battle, and one that humans can no longer fight alone,” Candela said. “AI-powered attackers are too powerful, too competent, too fast, and too numerous, meaning the only way to fight back is at the same pace and intensity.”

He emphasized that the product adapts to new malware and stays updated to match the current state of the most sophisticated attacks.

The Bigger Picture: AI vs. AI in Cybersecurity

Beelzebub’s approach reflects a broader trend in the industry. As attackers weaponize AI to automate their campaigns, defenders are being forced to respond with AI of their own. The days of relying solely on human analysts to spot and stop intrusions are fading fast.

The funding round also signals growing investor confidence in deception-based defense. It’s a niche but increasingly vital segment of the market, and Beelzebub’s hybrid red/blue team model gives it a distinctive position.

For those tracking the space, it’s worth watching how the company scales its threat intelligence network and whether its on-premises offering gains traction with NIS2-regulated firms.

Related coverage: AI-powered email security funding and composable security operations platforms have also drawn significant investment recently.

Malicious Vite npm Packages: A New Supply Chain Threat

Cybersecurity researchers at Checkmarx have uncovered a cluster of seven malicious npm packages targeting the Vite frontend tooling ecosystem. Dubbed ViteVenom, the campaign is an evolution of an earlier operation called ChainVeil, which used a four-tier blockchain-based command-and-control (C2) infrastructure spanning Tron,

The attackers are sneaking remote access trojans (RATs) into developer environments through packages that appear legitimate. If you’re a frontend developer using Vite, this is a wake-up call.

How the ViteVenom Attack Works

The malicious packages are designed to slip past standard security checks. They use blockchain transactions as their C2 channel, making detection far harder than traditional HTTP-based malware.

Checkmarx noted that ChainVeil’s C2 infrastructure was “unprecedented” because it relied on smart contracts to issue commands. ViteVenom continues that trend, embedding malicious code in packages that mimic Vite plugins or utilities.

The Seven Malicious Packages

While Checkmarx didn’t name all seven packages in public disclosures, the campaign targets developers who install Vite-related dependencies. The packages are likely published under names that resemble popular Vite plugins, a common typosquatting tactic.

Once installed, the RAT can steal credentials, exfiltrate source code, and even pivot to other systems on the developer’s network.

Why Blockchain C2 Is a Game-Changer for Attackers

Traditional C2 servers can be taken down by security teams. Blockchain C2, however, is decentralized. Commands are embedded in transactions on networks like Tron, making them nearly impossible to shut down.

This is a significant escalation in supply chain attacks. Security tools that rely on blocklists or domain reputation won’t catch this activity.

For developers, the risk is real: a single malicious package can compromise your entire project and your machine.

How to Protect Yourself from Malicious npm Packages

Here’s what you can do to stay safe:

• Audit your dependencies regularly with npm audit or tools like Snyk.
• Check package popularity and publish dates before installing. New packages with few downloads are red flags.
• Use lockfiles to pin exact versions and avoid surprise updates.
• Run scans for known malicious packages, especially those flagged by npm security advisories.
• Consider using a proxy registry that filters malicious packages.

Also, be cautious with packages that request broad permissions or include obfuscated code. If something looks off, inspect the code before running it.

What This Means for the Vite Ecosystem

Vite has become a go-to build tool for modern frontend projects, so it’s no surprise attackers are targeting it. The ViteVenom campaign shows that even trusted ecosystems aren’t immune.

Checkmarx’s findings highlight the need for stronger supply chain security. Developers should treat every dependency as a potential attack vector.

If you’ve installed any suspicious Vite-related packages recently, review your environment immediately. Remove unknown dependencies and rotate any credentials that might have been exposed.

The threat landscape is evolving, and blockchain-based C2 is just the beginning. Stay vigilant.