Introduction

Back in October 2022, Kaspersky GReAT was looking into a peculiar advertisement on a hacking forum. A cybercriminal had been advertising a UEFI bootkit, dubbed BlackLotus, that could bypass Secure Boot, for $5,000. Given the low price for such powerful capabilities, the cybersecurity community initially took this advertisement with a grain of salt. As no BlackLotus samples were found in the wild, no one could at the time verify the claims made in the forum post.

BlackLotus advertisement

However, about half a year later, in March 2023, ESET confirmed that the BlackLotus advertisement was genuine. The researchers tracked down samples of this bootkit and confirmed that it can indeed bypass Secure Boot through a vulnerability called Baton Drop (CVE-2022-21894). As this vulnerability was still exploitable in 2023, Microsoft had to release multiple security updates to address it.

Since the release of these updates, the BlackLotus bootkit has gone quiet — no publicly known attacks have used it. Yet in early 2026, we discovered a malware sample similar to BlackLotus that was used in an APT campaign in Central Asia. In this report, we share our findings, including new payloads and possible attribution of this campaign.

The return of the infamous bootkit

In early 2026 we identified a complex APT attack targeting diplomatic agencies in Central Asian countries — namely Tajikistan, Uzbekistan, and Turkmenistan. This attack involved a UEFI bootkit that we dubbed BlackX. Having examined it, we found it to be similar to BlackLotus.

Attackers have been deploying BlackX after compromising the victim machine networks and obtaining administrator privileges. To install the bootkit, we observed them dropping the following files:

• inst.exe, an installer executable
• PVMDIJGKAX2, an RC4-encrypted extensionless file containing files and strings required for the installation

The installer was launched with the command line C:\ProgramData\Intel\inst.exe PVMDIJGKAX2 Q46F6MDLI3RX2KV1. The last argument here is the RC4 key used for decryption.

Curiously, the file inst.exe does not contain any malicious strings or embedded blobs. All the files needed to install the malicious payload are inside PVMDIJGKAX2, which makes it harder for incident responders to retrieve the bootkit components.

The installer package drops a set of files to both system and UEFI partitions, designed to exploit the Baton Drop vulnerability. We noted that the implementation of the exploit is remarkably similar to the one used in the BlackLotus malware. As with BlackLotus, the goal of the exploit is to install custom Machine Owner Keys (MOK), which allows loading arbitrary EFI files on the target machines to manipulate Secure Boot.

While the vulnerability exploitation chain was implemented very similarly to BlackLotus, we spotted differences between the exploits in BlackLotus and BlackX. That said, the source code of the BlackLotus bootkit has never been published online. This could mean that the attackers behind BlackX either reverse-engineered the BlackLotus bootkit or purchased its source code from cybercriminals. Either way, this is an interesting observation, as we rarely see APT actors borrow ideas or code from cybercrime actors.

Payload launched by the bootkit

The UEFI payload delivered during the Baton Drop exploitation has a classic bootkit goal — to inject a malicious driver into the Windows kernel. This driver differs from the one deployed by BlackLotus and is signed with a certificate belonging to an organization named 温江区明宇网络技术服务工作室 (English: Wenjiang District Mingyu Network Technology Service Studio). At the time of writing, this certificate (serial number: 0bcdf5e0da32d68900b8b7638459d072) has been revoked by its issuer.

The driver aims to tamper with security solutions installed on the machine. For example, it disables process, thread, and registry callbacks related to security solutions. It also tampers with ETW providers so that security solutions cannot access their output. Curiously, some of the tampering procedures are designed to run only if Kaspersky or Avast solutions are installed on the infected machine. However, during dynamic analysis of this driver, we found that these procedures are useless — they fail to interfere with Kaspersky products in any way.

The driver also protects the bootkit’s files from tampering. It intercepts kernel requests to open these files and returns a handle to calc.exe instead of a malicious file handle.

User-mode second-stage payloads

Apart from tampering with security solutions, the driver also deploys a user-mode payload. We observed two such payloads, used for cyberespionage, in the wild. One of them is a RAT that can load further-stage .NET payloads. This RAT uses cloud storage for C2 server communications — we observed its versions using Google Drive, OneDrive, and Yandex.Disk. Having connected to a cloud storage service, this RAT iterates over files available there and interprets them depending on the file suffix. Below is a list of supported suffixes:

We also observed another payload delivered through the bootkit, which we dubbed OxiDLLRunner. It is a malicious DLL obfuscated with OLLVM. It is loaded by the legitimate file googleupdate.exe (original filename: jcef_helper.exe, MD5: AF8A983B9ADA07EAE1259BF2953ABD40), which is also delivered by the bootkit. The DLL decrypts the contents of the file v8_context_snapshot.bin, which resides in the same folder, and then jumps to the beginning of the decrypted content. The decrypted data starts with shellcode followed by a PE file, which is a Rust payload designed to communicate with the www[.]archivetj[.]com C2 server. During communication, this payload is supposed to receive DLL files exporting a function named create_plugin, which are then executed in separate threads.

Attribution

Based on the following facts, we have attributed this attack to the APT41 threat actor with medium confidence:

• In June 2026, ESET published a blog post on attacks by the FishMonger threat actor, which the community assumes to be related to APT41. This post describes a kernel implant with a PDB path starting with C:\Users\xdd\Desktop. While describing the campaign, the post also suggests the possible use of an unspecified UEFI bootkit component, which might exploit CVE‑2023‑24932. We observed the identifier “xdd” used multiple times in the BlackX bootkit payloads, so we assess that BlackX is the UEFI bootkit referenced in the ESET blog post;
• The payloads deployed by BlackX are loaded using a legitimate executable abused via DLL sideloading, and an encrypted file — this pattern is common among Chinese-speaking threat actors;
• The OxiDLLRunner file has been obfuscated with OLLVM, an obfuscator commonly used by Chinese-speaking actors.

Why did the target organizations get infected?

During our initial analysis of BlackX, we thought that attackers succeeded in deploying it due to the negligence of the target organizations. Baton Drop, the vulnerability used to deploy BlackX, was disclosed back in 2023 and was patched, so we first thought that the infected machines lacked the relevant updates. However, our investigation proved this wrong — the targeted machines were fully patched.

Instead, we found out that the problem was with the patch itself. As it turned out, the patch for the Baton Drop vulnerability is designed to revoke the “Windows Production PCA 2011” certificate used to sign vulnerable UEFI components. However, this revocation does not happen automatically upon patch installation — the user has to trigger it manually. This is done with the following steps:

• A registry flag is set inside HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Secureboot;
• A scheduled task named Secure-Boot-Update is created.

This patch is designed to be applied manually — for instance, applying it automatically could break dual-boot systems relying on the “Windows Production PCA 2011” certificate to load. At the same time, this patch deployment practice creates a problem — as it is extremely unusual for a system administrator to take extra steps after installing an update. Usually, when applying patches, system administrators take no further action unless an installation failure occurs. These instructions for activating the patch are buried deep on the Microsoft website, and we assess that they are quite hard for an average IT administrator to notice.

Curiously, Microsoft’s plan for deploying the Baton Drop patches mentions an “Enforcement Phase” that involves revoking the “Windows Production PCA 2011” certificate automatically, without user interaction. At the time of publishing this post, this stage has not been implemented. On October 19, 2026, this certificate expired, but this does not prevent the vulnerable binaries from loading unless the certificate is blocklisted.

It is likely that the threat actor behind this campaign took into account that few organizations realized the importance of manually activating the Baton Drop patch. As such, we believe they decided to invest time in understanding how BlackLotus works and creating a copycat of this bootkit.

Conclusion

Patch management is important for every organization, as updates from vendors help protect software and hardware against known vulnerabilities. However, sometimes installed patches fail to protect against the flaws they address. The BlackLotus story we described is not the only such example — for instance, in 2025 Microsoft released an update against the ToolShell vulnerability that also had to be applied manually.

However, modern security solutions are designed to account for such cases when installed patches fail to protect against security flaws for one reason or another. For instance, the Exploit Prevention component available in Kaspersky products prevents exploits from running if the patch is not applied, or if it is a zero-day vulnerability with no available patch. It is thus crucial to equip all machines in an organization with security solutions and constantly monitor the alerts they raise.

Cyble scanned 1.3M+ newly registered domains and found AI-built sites leaking live secrets to the browser: OpenAI keys, Stripe keys, and database tokens.

Spinning up a website no longer takes a developer. A prompt to an AI app-builder returns a working, styled, deployable site in minutes, and when one template works, the same operator clones it across a dozen domains before launch. That is also how secrets end up in plain sight.

Over a month, Cyble Research and Intelligence Labs (CRIL) scanned more than 1.3 million newly registered domains, identified the ones built by AI app-builders, and read their public front-end code for exposed credentials. We built our own classifiers to spot AI-generated websites as they are registered, paired with a scanner that pulls exposed keys and credentials straight from their public code. On 35 of the 22,306 AI-built sites we identified, a genuinely private secret was sitting in the browser: live OpenAI keys, Stripe secret keys, backend database tokens, and, in several cases, a single credential cloned across a whole fleet of domains.

Thirty-five is a small number, since counting every key-shaped string on an AI-built site is easy and misleading, as most are meant to be in the browser; a single reCAPTCHA key appeared on thousands of sites. We kept only the credentials that are never supposed to leave a server. Those are the findings below.

This report walks through the two leak patterns we observed, what appears exposed, what an attacker can do with these keys, and how to shut down the exposure.

AI-built Websites are leaking the keys that power them

An AI-built site generates its content with a Large Language Model (LLM), and the key to that model often ships in the page the visitor downloads. That is the plainest pattern we found, and the most on-the-nose.

Five AI-built sites carried live OpenAI API keys in their client-side code, sixteen distinct keys across them, and four of the five are confirmed AI-generated. One AI-writing site stands out on its own: it shipped twelve different OpenAI keys, one per content category, as plain values in its bundle. Figure 1 shows a single entry, with the key masked.

A handful of sites, but each key is sensitive to exposure. An OpenAI secret key is a billing credential: anyone who reads the page can spend the owner’s balance, exhaust the rate limit, or run the bill up until someone notices and revokes it. Twelve keys in one bundle are twelve live credentials shipped at once, and five sites like this are five accounts an attacker can exploit.

Clone-and-leak: one secret across a whole fleet

An AI builder makes the hundredth site as cheap as the first, so an operator clones a working template across many domains, and whatever secret sat in the template now sits on every copy. We saw this in two shapes.

A Slack webhook shared across a 13-domain fleet. Thirteen near-identical domains, each a short word (ask, call, chat, get, hello, hey, hi, its, join, let, meet, my, see), glued to one brand, all served the same client bundle, and all embedded the same live Slack incoming-webhook URL. Figure 2 shows the URL in the shipped bundle, masked.

Anyone who can read the page can post messages to the operator’s Slack workspace via that webhook. That works for spam, for phishing aimed at the operator’s own staff, or for flooding a channel.

One Stripe secret key across four domains and two brands. A single Stripe secret key appeared on four domains under two separate brand names. The same sk_live_ value across all four points to one operator, or one shared Stripe account, behind every site. Figure 3 shows the assignment.

A Stripe secret key provides full API access to the account, including the ability to create charges, issue refunds, and read customer and payout records. That one key exposes the operator’s whole payment account on every domain in the fleet.

The 13-domain Slack fleet runs on Bubble, a no-code and AI app builder, as confirmed by Bubble’s x-bubble-perf response header and client markers. Bubble serves each app from a dynamic.js bundle, and any value a developer places in a client-accessible field ships inside that bundle in plain text. The Stripe cluster runs on a Next.js build instead, which shows that the pattern spans builders rather than tracing to a single platform.

What else is exposed

Four AI-built sites exposed a Supabase service role JSON Web Token (JWT) on their front ends. A service-role token bypasses row-level security and grants full read and write to the project database. In terms of impact, it sits at the top of the set because it opens the entire backend rather than a single service.

Most of the remaining secret-looking strings on AI-built sites carry no/minimal exposure.

• 3,824 sites shipped browser-safe values: reCAPTCHA site keys, Stripe publishable keys, Supabase anon keys, builder telemetry tokens.
• 2,172 sites carried Google API keys, which matter only if they are unrestricted.
• 1,461 sites exposed a Supabase or Firebase backend address, which is a real exposure only when that backend’s rules are open.

Filtering those out is why the confirmed count is 35, not several thousand. Table 1 breaks the 35 down by type. The values stay withheld, and a few sites appear under more than one row.

Table 1 – Confirmed private-credential exposures across 22,306 AI-built sites, by type.

What an attacker can do with these

Every credential in Table 1 is a working key to a live account. An attacker reads a public web page, copies a value, and calls an API. Here is what that buys them, key by key.

• OpenAI and other LLM keys: Run inference on the victim’s account until the bill or rate limit stops them. On a metered account, that is direct financial loss; on a funded one, it is a free, anonymous model endpoint that the attacker can point at anything, billed to the victim and running under their quota and identity.
• Stripe secret key; Drive the victim’s payment account through Stripe’s own API: create and refund charges, push the account toward fraud, and pull the customer list, email addresses, and payout history. One key is a money problem and a data-breach problem at once.
• Supabase service-role token: Read and write the entire production database, past whatever row-level security the app configured. That means stealing every user record, altering or deleting data, and planting content the app will serve back to its own visitors.
• Slack incoming webhook:  Post into the operator’s own Slack as if from inside. Flooding a channel is a nuisance case; the real risk is internal phishing.
• AWS access key: With the paired secret and enough permission, reach into the owner’s cloud: spin up resources (crypto-mining is the usual first move), read storage buckets, or pivot further in.
• GitHub token or OAuth secret: Read or push to the owner’s source repositories or misuse their identity integration.

One exposure counts many times over. When the key sits on a cloned fleet, it is live on every domain in that fleet, so an attacker who finds it on one site has compromised all of them.

Conclusion

AI app-builders made it almost free to publish a website, and just as free to clone one, and most of what they produce is worth nothing to steal.

The exceptions are what we went looking for. A small, steady set of these sites hand over the key that runs them, and when an operator copies one site across a fleet, the leak rides along to every domain.

Against the millions of domains registered each month, the number is tiny, but every hit is a live account an attacker can reach and exploit just by opening the page.

If you build on these tools, read your own shipped code before you publish it. If you defend an organization, assume your keys may surface on sites you never built and look for them.

Cyble’s Threat Intelligence Platforms continuously monitor emerging threats, infrastructure, and activity across the dark web, deep web, and open sources. This proactive intelligence empowers organizations with early detection, impersonation, infrastructure mapping, and attribution insights. Altogether, these capabilities provide a critical head start in mitigating and responding to evolving cyber threats.

Our Recommendations

For developers and AI-builder users:

• Keep every secret key (OpenAI sk-, Stripe sk_live_, a Supabase service-role token) in a server-side function or backend workflow, never in a client-accessible field.
• Use only the browser-safe variants on the front end, publishable and anon keys, and enforce access with Supabase row-level security, Firebase rules, and API key restrictions.
• Rotate any key that has ever sat on the client side, and treat it as already compromised.
• Read your own page source before you ship, and search it for sk-, sk_live_, service_role, and hooks.slack.com. If you find them, so can anyone.

For SOCs and defenders:

• Monitor public code and newly registered domains for your own keys and webhooks; don’t just paste sites.
• Treat a secret found on one domain of a template fleet as exposed on all of them, and scope the response to the fleet, not the domain.

The post The Keys Are in the Browser: Secret Exposure Across AI-Built Websites appeared first on Cyble.

Researchers at Hudson Rock found that cybercriminals hijacked HBO Max’s verified Reddit account and used it to run 108 malicious ads over roughly 48 hours.

The ads used HBO Max’s trusted corporate account to promote fake AI tools, developer software, and macOS utilities, lowering potential victims’ guards.

Some ads directed users to convincing HBO lookalike sites that claimed to offer a native HBO Max app for macOS or a promotional download. But instead of providing an installer, the sites instructed visitors to open Terminal on their Mac or, on Windows, the Run dialog or PowerShell, and paste in a command.

This is the hallmark of a growing social engineering technique known as ClickFix.

ClickFix attacks disguise malicious instructions as a routine technical step, such as fixing an error, completing a CAPTCHA, verifying that you are human, or installing software. A web page may silently copy a command to the clipboard, then guide the victim through pasting and running it, by which they will infect their own device.

Researchers at ADAMnetworks have dubbed the operation behind the HBO Max ads “PasteSwitch.” Its infrastructure appears to tailor the next stage to the visitor’s device and the lure being used.

Observed macOS payloads included MacSync and AMOS infostealers designed to steal browser credentials and profiles, Telegram data, Apple Notes, saved passwords, and cryptocurrency wallet recovery phrases.

Windows users could end up with the Amatera infostealer, which runs in memory. The operation has also been linked to cryptocurrency clipboard hijackers, which monitor copied wallet addresses and replace them with an attacker-controlled address before a transaction is sent.

How to stay safe

Reddit admins paused the ads and opened a security investigation after reports came in, but it is important to remain vigilant. Reportedly, ClickFix was responsible for more than half of all malware loader activity in 2025.

One reason for its success is that campaigns continue to add new methods for tricking users and different commands for avoiding detection.

Users of macOS Tahoe 26.4 or later may be warned when pasting text into Terminal, but it doesn’t appear for everyone, so you shouldn’t rely on that alone.

Malwarebytes can provide additional protection at several stages of a ClickFix attack. Browser Guard warns when a website tries to copy something to your clipboard, web protection blocks known malicious sites, and real-time protection can detect malware delivered by the campaign.

With ClickFix running rampant and inventing new methods all the time, it’s important to be aware, careful, and protected:

• Treat ads with caution. A verified account does not guarantee that an ad is safe. Visit the company’s official website directly instead of downloading software through an advertisement.
• Slow down. Don’t rush to follow instructions on a webpage or prompt, especially if it asks you to run commands on your device or copy-paste code. Attackers may use countdowns, user counters, or other pressure tactics to make you act quickly.
• Don’t run commands from untrusted sources. Never run code or commands copied from websites, emails, ads, or messages unless you trust the source and understand what the command does. Check the instructions against official documentation or contact the company’s support team.
• Secure your devices. Use an up-to-date, real-time anti-malware solution with a web protection component. Malwarebytes blocks ember-bridge.com, which is part of the PasteSwitch infrastructure.
  
• Educate yourself on evolving attack techniques. Understanding that attacks may come from unexpected vectors and evolve helps maintain vigilance. Keep reading our blog!

From reporting threats to removing them.

Cybersecurity risks should never spread beyond a headline. Keep threats off your devices by downloading Malwarebytes today.

Grindr has reportedly agreed to pay £26 million (around $35 million) to settle a UK privacy lawsuit alleging that it shared sensitive user data, including some users’ HIV status, with advertisers.

The claim was brought by London law firm Austen Hays on behalf of roughly 12,000 UK Grindr users. It alleges that the dating app breached privacy and data-protection laws during a period ending in early 2020.

The claimants allege that Grindr shared personal and highly sensitive information with advertising companies without consent. According to Austen Hays, the shared data may have included ethnicity, HIV status, the date of a user’s last HIV test, and whether they used pre-exposure prophylaxis (PrEP).

At the time of the alleged data sharing practices, Grindr was owned and controlled by the Chinese gaming company Beijing Kunlun Tech. Grindr was sold to US owners in 2020.

According to a US regulatory filing, Grindr will make two payments of £13 million: one by December 31, 2026, and the second by March 31, 2027.

In its SEC filing, Grindr said that the settlement is not an admission of liability and, while it disputes the allegations, it:

recognizes and acknowledges the distress and loss of trust expressed by some of its UK users regarding that pre-2020 period.

The UK settlement follows a separate enforcement case in Norway. The country’s Data Protection Authority found that Grindr had shared users’ personal data with advertising partners for behavioral advertising without a valid legal basis.

These cases illustrate a crucial privacy point: information does not need to be explicitly labeled as medical information or information about sexual orientation to expose intimate details about someone. Advertising identifiers, IP addresses, locations, device information, and confirmation that a person uses a particular app can be combined to identify them or draw sensitive conclusions about their life.

Many free apps rely on advertising SDKs, analytics providers, and other third parties to make money. These integrations can receive identifiers and event data that help target or measure advertising, but they can also create extensive trails of user behavior.

How to protect your privacy on dating apps

Grindr says it has overhauled its privacy program since 2020 and remains committed to user control and responsible data practices. Even so, dating apps can hold unusually personal information about their users.

To limit what you reveal:

• Review the app’s privacy settings and turn off optional personalized advertising where available.
• Limit your profile to details you’re comfortable sharing with potential matches.
• Avoid linking a dating profile to public social-media accounts unless you want identities to be easily connected.
• Revoke location permissions when you’re not actively using the app, or choose “while using the app” rather than continuous access where your operating system offers it.
• Keep the app, your operating system, and your security software updated.
• Watch for romance scams and extortion attempts, particularly requests to move the conversation off the app, send money, share intimate photos, or reveal identifying information.

If you’re unsure whether a message may be part of a scam, you can check it with Malwarebytes Scam Guard, which can help you assess the conversation and decide what to do next.

Scammers don’t need to hack you. They just need you to click once. 

Malwarebytes Identity Theft Protection catches suspicious activity before it becomes a problem.

While monitoring Mirage Kitten activity, we uncovered a previously undocumented malware family that we dubbed NodeRabbit. We identified the first sample on a system in Afghanistan. Further threat hunting revealed two additional, more advanced, variants: one on a system in Egypt and another on a system in Ethiopia.

NodeRabbit is a cross-platform remote access trojan (RAT) built with Node.js. It targets Windows, Linux, and macOS. Its operators deliver it through spear-phishing messages on LinkedIn and other job search platforms that contain trojanized coding challenge archives.

During the same investigation, we discovered another previously undocumented malware family that we dubbed PollCat. Like NodeRabbit, PollCat is a cross-platform RAT, but it is written in obfuscated JavaScript also distributed through trojanized coding challenge archives.

Mirage Kitten has historically relied on native malware written in languages such as C, C++, and Go, often deploying it through DLL search-order hijacking. NodeRabbit and PollCat represent the first publicly documented use of Node.js- and JavaScript-based malware by this APT group.

Kaspersky’s products detect this threat as Trojan.JS.MirageKitten.*

Background

During recent threat research, we detected suspicious activity on a system in Afghanistan. We traced it to an archive containing a software development project that the user may have received during a job application process. The archive purported to contain a coding challenge for candidates applying for an engineering role.

The archive, Front-Technical-Challenge.zip (MD5: 1EA83E4E4592B01E4ACAB63EB867BEE5), was hosted in an Amazon S3 bucket at: https://oracle-challenge.s3[.]us-east-1.amazonaws[.]com/Front-Technical-Challenge.zip

It contained TaskFlow, an app for software engineering assessment built with Express, React, and Vite. The accompanying README instructed the candidate to review the application and fix defects in its frontend. It also claimed that server.js was bug-free and should not be modified, conveniently directing attention away from the only application source file the attackers had altered.

README file for a trojanized coding challenge app

The README also imposed a three-hour time limit and prohibited the use of AI assistants. Notably, an AI code-review assistant tasked with auditing the project would likely have flagged the suspicious first-line import of an unknown npm package and warned the targeted developer that the project was trojanized.

Rules and time limit included in the trojanized coding challenge app README file

The first line of server.js imported a trojanized npm package named colorized_terminal, version 2.1.0. The attackers bundled the package directly in the challenge task archive’s node_modules directory rather than publishing it to the npm registry. When imported, the package silently launched an implant from node_modules/.cache/.320697f1/index.js as a detached background process.

Retrospective threat hunting across our telemetry revealed the broader scope of the campaign. We identified three NodeRabbit variants with a shared code lineage; each was recovered from a system in a different country. The operators delivered the variants through similarly themed coding challenges and used two trojanized packages, colorized_terminal and pretty-log, both pinned to version 2.1.0.

The campaign also delivered PollCat, a second RAT with a substantially different structure, through a separate coding challenge lure. We’ll analyze PollCat later in this research.

Initial access

The infection chain begins with fake recruiter accounts contacting prospective targets on a job search platform. According to a publicly cited source, a threat actor posing as a talent acquisition specialist at a major technology company contacted a software engineer and advertised a job opening, inviting the target to complete a technical assessment.

The target received a link to a coding challenge hosted on Amazon S3 and was pressured to download and run the project immediately. This public post matches the delivery chain we reconstructed from our telemetry: recruiter outreach on a job search platform, a coding challenge presented as a technical assessment, and a trojanized project archive hosted on legitimate cloud infrastructure.

NodeRabbit RAT: the first variant

We discovered the first NodeRabbit variant on a system in Afghanistan. The malware was concealed within the TaskFlow assessment at node_modules/.cache/.320697f1/index.js and executed by the trojanized colorized_terminal package.

Once running, NodeRabbit generates a unique agent identifier from available host information. It calculates the SHA-256 hash of the hostname, username, operating system version, architecture, and MAC address, then truncates the result to its first 32 hexadecimal characters.

NodeRabbit binds a TCP listener to 127.0.0.1:48739. This listener acts as a single-instance mechanism. If the malware cannot bind to the port, it assumes that another instance is already running and terminates silently.

NodeRabbit uses a persistence mechanism for each operating system:

The malware communicates with its command-and-control servers through three API endpoints, choosing from the following Azure-hosted C2 infrastructure addresses. On failure, it switches to the next C2 address:

1.	https://plugplay.azurewebsites[.]net
2.	https://Rgbteller.azurewebsites[.]net
3.	https://Wslwebui.azurewebsites[.]net

NodeRabbit serializes each C2 request object as JSON and wraps it with AES-256-GCM. The AES key is the SHA-256 digest of an ASCII seed embedded into the agent. Every request uses a fresh 12-byte IV and a 16-byte authentication tag:

The malware sends encrypted requests using the following structure:

{
  "d": "base64(IV || ciphertext || authentication_tag)",
  "_r": "8 hexadecimal characters",
  "_t": "epoch timestamp"
}

C2 responses are structured the same way and may contain a command to execute. We observed the first NodeRabbit variant supporting 11 commands:

NodeRabbit RAT: the second variant

Retrospective threat hunting following the discovery in Afghanistan led us to a second infection on a system in Egypt. This sample is a more advanced NodeRabbit variant, launched through the trojanized pretty-log package instead of colorized_terminal.

Before running its core functionality, the malware checks whether the host resembles an analysis environment. It terminates if it detects limited system memory, a low CPU count, short system uptime, analyst-associated usernames or hostnames, or common analysis tools running on the system.

Before terminating, the malware generates benign HEAD requests to www.google.com, www.microsoft.com, and www.cloudflare.com, then exits without ever contacting its C2 infrastructure. Most likely, it attempts to look less suspicious by showing some benign activity before exiting.

Variant 2 implements partial corporate proxy support: it checks HTTP(S) proxy environment variables, Windows Internet Settings, including an explicit PAC URL, and WinHTTP configuration; tunnels its HTTPS C2 through HTTP CONNECT. It first tries to establish an unauthenticated connection. If it fails, it retries using URL-embedded basic credentials. Finally, it delegates Windows NTLM/Negotiate challenges to curl.exe --proxy-anyauth --proxy-user. It caches the proxy-discovery result, including when no proxy is found, for five minutes. If the polling loop detects a network-interface or IP-address change, it clears the cache and runs proxy discovery again on the next checkin.

To make sure a single instance is running, Variant 2 uses a host-specific port derived from the agent identifier instead of the fixed TCP port used by the first variant. It interprets the first four hexadecimal characters of the identifier as an integer and applies the following calculation: 41984 + (value mod 5000).

The resulting listener port falls between 41984 and 46983. Unlike the shared port used by Variant 1, this port varies depending on the infected host.

For persistence, Variant 2 masquerades as Intel Driver & Support Assistant. The exact persistence mechanism, once again, depends on the operating system.

NodeRabbit RAT: the third variant

Further threat hunting identified a third NodeRabbit variant on a system in Ethiopia. Like the second variant, it is launched through the trojanized pretty-log package. It retains much of the previous variant’s functionality but introduces significant changes to its command-and-control configuration, command set, and persistence mechanisms.

The third variant communicates with its C2 infrastructure through a different set of API endpoints:

We observed the malware using a C2 chain composed of Azure- and Cloudflare-hosted domains.

1.	https://visitfinancedentists[.]com
2.	https://kyrasey-f8hfexa5cqamh7fk.westeurope-01.azurewebsites[.]net
3.	https://healthcomfsdpower[.]com

For persistence, Variant 3 implements the following mechanisms depending on the operating system in use:

A new command, agent:servers, replaces the active in-memory C2 server list and can write the updated list to .sv.json. The third variant retains the original 11 commands and adds 12 new ones, bringing the total to 23.

Beyond the persistence mechanisms described above, Variant 3 introduces two additional persistence mechanisms that relaunch the malware through common developer workflows.

1. Malicious VS Code extension

The persist:vscode command first copies the payload to its build-specific install path. If a compatible extension directory exists, it creates a fake extension displayed as GitHub Copilot Helper, with the description AI coding assistant helper service and the activation event on StartupFinished.

The extension’s extension.js file attempts to start the installed payload as a detached Node.js process. To look less suspicious to the user, it uses a trusted publisher name borrowed from local extension metadata or a trustedPublishers value found in state.vscdb. However, no signature or trusted status is copied.

Separately, the handler tries to disable Workspace Trust if the VS Code User directory exists. On Windows, it attempts to establish persistence using a current-user Run registry key value even if the extension directory is missing.

2. Git hook injection

Git-hook persistence works in two steps. First, persist:projects:scan checks recent VS Code workspace paths directly. Under common locations such as ~/projects and ~/source, it checks only the first 60 immediate children, not the root itself, and returns no more than 20 repositories.

For a selected repository, persist:project:inject appends a marked launcher to .git/hooks/post-merge and .git/hooks/post-checkout by default. The marker is # shepherd-persist; the line following the marker attempts to start the installed payload with Node in the background. A later Git operation must trigger one of those hooks, and the referenced Node executable and payload must still exist.

PollCat RAT

While tracking NodeRabbit infections, we discovered another malicious tool we dubbed PollCat, which is also distributed under the guise of a programming challenge. The sample we obtained resides inside RankChallenge-react, a React code-fixing challenge presented as a time-limited developer assessment. Running the project invokes npm i && node index.js, which starts the local application and attempts to open the challenge in the user’s browser.

Although the visible exercise is not a security CTF, the project uses CTF terminology in several places. The root package is named ctf-server, the backend prints CTF server running, the frontend uses several ctf-* storage keys, and the tutorial refers to path/to/ctf. These repeated labels, together with instructions that do not fully match the delivered application, are consistent with an AI-assisted or template-generated project. One possible explanation is that the attacker prompted an AI coding assistant to create a CTF-style React platform and later inserted the malicious components.

README instructions and challenge overview included in the trojanized React coding project

The PDF tutorial contained in the same archive as the project tells the target to click Continue, enter a six-digit OTP code, and complete the challenge within a one-hour session. It states that codes are supplied by the recruiter, are single-use, and expire quickly; the visible login page also claims that codes rotate every 30 seconds. In the delivery scenario described by the investigation, the threat actor posing as a recruiter could provide the code directly to the targeted developer. This gives the operator control over access to the lure, while the expiring code and countdown create a sense of urgency, pressuring the target to run the project and complete the assessment quickly, potentially accelerating the infection process.

One-hour session window enforced by the trojanized coding challenge

The bundled .env file contains the JWT signing secret, OTP service URL, and OTP client ID.

Configuration embedded in .env file of the trojanized coding project, including the OTP service URL and client identifier

The application forwards submitted codes to an attacker-managed domain registered in late June-2026: https://lifespotify[.]com/api/users/b879746e-fed9-4211-a6da-4d8223681267/otp/validate.

That said, PollCat starts independently of the OTP authentication process. During application startup, app.js loads requireAuth.js, which imports and immediately starts the malicious requireObjects.js component. PollCat can therefore begin C2 registration and command polling while the application is still loading, before the user enters an access code.

A failed OTP validation prevents the user from accessing the protected challenge features, but PollCat continues running in the background. A successful OTP validation issues a JWT and creates another worker that starts an additional PollCat instance. The first authenticated request also triggers the persistence attempt.

Persistence starts when the first request carrying a valid JWT reaches the protected middleware. PollCat then uses one of the following methods:

Once active, PollCat identifies the host as 129--<hostname> and iterates over the following C2s until registration succeeds:

1.	https://sahi-finance[.]com
2.	https://GamebarAppinformation[.]azurewebsites[.]net
3.	https://GamebarApp[.]azurewebsites[.]net

To register, it sends the following HTTP request to the C2:

POST /beacon HTTP/1.1
Host: <c2-host>
Content-Type: application/json

{"clientId":"<client-id>","type":"poll","pcName":"<hostname>","userName":"<username>"}

On successful registration, PollCat expects an unusual HTTP 400 response containing a socket identifier and optional timing values:

HTTP/1.1 400
Content-Type: application/json

{"socketId":"<socket-id>","pollInterval":<poll-interval-ms>,"jitterTime":<jitter-ms>}

After registration, PollCat sends host information to /gate/hello, polls /gate/fetch for commands, and returns results through /gate/submit. All endpoints in use are presented in the table below.

By default, PollCat RAT polls every two minutes with up to five seconds of jitter. Commands and results are stored as little-endian binary records and carried as Base64 text.

PollCat RAT declares 22 commands, but three of them have no implementation:

The command names UPLOAD, DOWNLOAD, and CHUNKED_DOWNLOAD are written from the C2’s perspective. UPLOAD sends a C2-hosted file to the victim’s machine, while the two download commands transfer victim files back to the C2.

EVAL_JS runs JavaScript supplied by the C2 and gives that code access to Node.js modules, files, processes, networking, and child-process functions.
SYSTEM_CHECK collects the names of running processes and lists files and folders from:

• %SystemDrive%\Program Files
• %SystemDrive%\Program Files (x86)
• %LOCALAPPDATA%
• %LOCALAPPDATA%\Programs
• %APPDATA%
• %USERPROFILE%
• %APPDATA%\Microsoft\Outlook
• %LOCALAPPDATA%\Microsoft\Olk\Attachments
• %USERPROFILE%\Documents

It also searches for folders matching 24 hardcoded strings corresponding to security software vendor names: ‘Google’, ‘Microsoft’, ‘Palo Alto Networks’, ‘Cisco’, ‘VMware’, ‘Fortinet’, ‘Citrix’, ‘CheckPoint’, ‘Juniper Networks’, ‘LogMeIn’, ‘Sophos’, ‘Symantec’, ‘Trend Micro’, ‘McAfee’, ‘Kaspersky Lab’, ‘ESET’, ‘Bitdefender’, ‘Avast Software’, ‘CrowdStrike’, ‘SentinelOne’, ‘Malwarebytes’, ‘BraveSoftware’, ‘Tencent’, and ‘Naver’.

When PollCat finds a matching folder, it lists that folder’s root contents. It does not recursively scan the entire product directory. The detailed inventory, including process names, directory listings, and collected paths, is sent as JSON to POST /api/system-details/result.

Infrastructure

Mirage Kitten continues to rely on Azure Websites and Cloudflare-backed domains to hinder infrastructure discovery and tracking. More importantly, the use of Microsoft Azure subdomains for C2 helps the traffic blend into legitimate organizational network activity. In some cases that we encountered during our research, the actors even incorporated the targeted organization’s name into the Azure subdomain, making C2 communications appear more like normal business traffic originating from an employee machine during regular business days.

Based on our analysis of Mirage Kitten’s infrastructure, we identified certain patterns across several command-and-control channels, including msmanagementgrp[.]com and visitfinancedentists[.]com

Further investigation based on these patterns led to the discovery of approximately 11 additional infrastructure assets attributed to the same group.

Victims

Based on our telemetry, we identified victims in fintech, aviation and aerospace sectors across the Middle East and Africa – specifically, in Egypt, Ethiopia and Afghanistan.

We also observed submissions of ZIP archives with trojanized projects containing NodeRabbit and PollCat to an online multi-scanner originating from several countries, including India, Türkiye, Israel, Iraq, Germany, and Ireland.

Attribution

We attribute this activity to Mirage Kitten with a high degree of confidence based on the following observations:

1. Structural similarities with the Retrograde/MiniFast native DLL backdoor (MD5:810F8E3B88EB05F710C09552941D6F56)
   • Initial C2 handshake and session establishment logic. Both PollCat and Retrograde/MiniFast follow a similar C2 handshake flow. Each builds a JSON request body containing host information and sends it via an HTTP POST request. Notably, both treat HTTP 400 as a successful handshake response rather than an error, parsing the response body to extract a socketId, which is then stored and used as the session token for subsequent C2 communication.
     

     Similar C2 handshake and socketId session establishment logic in MiniFast/Retrograde and PollCat

   • Host registration. Both PollCat and Retrograde/MiniFast register the infected host with the C2 server by sending a structurally similar JSON request body containing the session token and host information.
     

     Malware	Host registration request body	C2 endpoint
     PollCat	{“token”:”<socketId>”,”pcName”:”<host>”,”userName”:”<user>”,”domainName”:”<domain>”,”os”:”<os>”,”isElevated”:false}	/gate/hello
     MiniFast/Retrograde	{“token”:”<socketId>”,”pcName”:”<host>”,”userName”:”<user>”,”domainName”:”<USERDOMAIN>”,”isElevated”:<bool>}	/agent/init

   • Command fetching similarities. The similarities extend to command retrieval. Both PollCat and Retrograde/MiniFast periodically poll the C2 server using an HTTP GET request containing the previously assigned socketId as a token. Retrograde/MiniFast uses GET /agent/poll?token=<socketId>, while PollCat follows the same pattern with GET /gate/fetch?token=<socketId>, demonstrating a closely aligned C2 communication structure.
   • Beacon timing similarities. PollCat and the Retrograde/MiniFast share identical beacon timing defaults: a polling interval of 120,000 ms (0x1D4C0), a jitter of 5,000 ms (0x1388), and a retry timeout of 60,000 ms (0xEA60). This further highlights the structural similarities between the two C2 communication implementations.
   • Command set similarities. PollCat and Retrograde/MiniFast share several commands and command IDs. Notably, PollCat declares REQUEST_ELEVATION (0xB0) and PERSIST (0xB1) but does not implement them. In MiniFast, both are functional: 0xB0 performs UAC elevation, while 0xB1 creates the WindowsSecurityUpdate scheduled task for persistence.
     

     Command set similarities between MiniFast/Retrograde and PollCat, including shared command identifiers

   • Proxy authentication similarities. NodeRabbit delegates corporate-proxy NTLM/Negotiate authentication to curl.exe --proxy-anyauth --proxy-user, using the victim’s logon session. Retrograde/MiniFast native DLL implements the same approach natively through WinHttpQueryAuthSchemes and WinHttpSetCredentials with NULL credentials. This shared proxy-aware C2 design suggests the same development approach across both malware families.
2. Speaking of victimology, the attacks are consistent with Mirage Kitten’s known geographic targeting, with the group maintaining a strong focus on entities across Africa and the Middle East, this time with a particular focus on the aviation and FinTech sectors.
3. As for the operational infrastructure, Mirage Kitten has historically hosted its initial ZIP lures on legitimate third-party services. Previously, it used onlyoffice.com for this purpose. In this activity, the group shifted to Amazon S3 buckets.
4. Finally, the combination of Azure Websites and Cloudflare‑backed domains has been a hallmark of Mirage Kitten’s TTPs, which we have observed across NodeRabbit and PollCat.

Conclusions

Mirage Kitten’s latest activity marks a notable evolution in the group’s tooling: NodeRabbit and PollCat are the group’s first Node.js/JavaScript-based implants, departing from its usual native malware deployed through DLL search-order hijacking. The shift to cross-platform scripting gives the operators a single codebase that runs on Windows, Linux, and macOS, with payloads that blend naturally into developer workstations.

The delivery mechanism, however, remains consistent with Mirage Kitten’s historical tradecraft: the use of recruiter personas on LinkedIn to target critical sectors across the Middle East and Africa for cyberespionage purposes. We continue to track the group’s activity and will report on new developments in future publications.

Indicators of compromise

Additional IoCs are available to customers of our Threat Intelligence Reporting service. For more details, contact us at intelreports@kaspersky.com.

File hashes

CBAAF0900A13F28E380F49ADECEC932C  FrontEnd-Task.zip
1EA83E4E4592B01E4ACAB63EB867BEE5  Front-Technical-Challenge.zip
366515822D5AC1CC500711EF57A2E32E  Task-FullStack.zip
CF449F1992C2819E62AC44A0B06AC2E7  fullstack-1536.zip
E95A4366686E3F786EA3C056FAB5B0DA  webapp76592.zip
DE5AF16A3757EF700B01DC34D67079AE  webapp76531.zip
BE086789568441D0D7E4679AEE51F566  challenges-17831.zip
E259C5EDF158AAC4CFE14F77DDD0B196  challenges-17832.zip
291AC3ABE73C5158E59A437B75D5F0AA  Project-1802.zip
0962F56D7EC69F4F2A0162DCBE22116B  Case-34234.zip
795E053A990A1569FFDCB57F48F6D085  RankChallenge-react-6uJSX3-main.zip

Domains and IPs

oracle-challenge.s3[.]us-east-1.amazonaws[.]com
naturalapplication.azurewebsites[.]net
retaildemo.azurewebsites[.]net
tubitak.azurewebsites[.]net
rgbteller.azurewebsites[.]net
wslwebui.azurewebsites[.]net
plugplay.azurewebsites[.]net
crossdwm.azurewebsites[.]net
wdisystem.azurewebsites[.]net
wslmenus.azurewebsites[.]net
dnshnsdev.azurewebsites[.]net
hpjumpsrv.azurewebsites[.]net
storview.azurewebsites[.]net
healthcomfsdpower[.]com
visitfinancedentists[.]com
kyrasey-f8hfexa5cqamh7fk.westeurope-01.azurewebsites[.]net
greenyjsgfd.azurewebsites[.]net
helptellerbls.azurewebsites[.]net
timedrv.azurewebsites[.]net
userwellgtfs.azurewebsites[.]net
hecowime-aqdphyd4bbdef6es.westeurope-01.azurewebsites[.]net
msmanagementgrp[.]com
msmanagementgrpmedia[.]com
lifespotify[.]com
gamebarapp.azurewebsites[.]net
gamebarappinformation.azurewebsites[.]net
sahi-finance[.]com
healthful-hub[.]com
neumedicahealthcare[.]com
optimumhealthcredit[.]com
healthfullyrecipes[.]com
Refreshhealthandwellness[.]com
healthvitalitycare[.]com
aceofspadesmanagement[.]com
glmediaagency[.]com
digimediaskill[.]com
healthyweightplan[.]com
mens-health-online[.]com

The vulnerability landscape shifted significantly in Q2 2026. First, the number of registered CVEs reached an unprecedented level. This is driven primarily by the widespread adoption of AI, both for application development and search for security flaws. This resulted in entire new classes of vulnerabilities emerging, particularly in the Linux networking subsystem.

Second, security researchers have been publishing exploits for unpatched vulnerabilities more frequently. Publications like these can generate significant fallout, since they potentially open the door for attackers to target unprotected systems.

Statistics on registered vulnerabilities

This section provides statistical data on registered vulnerabilities. The data comes from Kaspersky’s vulnerability knowledge base, which draws on the CVE database as well as the Russian BDU database and GitHub Advisory (GHSA). As a result, the figures for previous reporting periods may differ from those published in earlier reports.

We examine the number of registered vulnerabilities for each month over the last five years. As the chart below shows, this number continues to surge, a trend reflected across all the databases we track. It’s driven primarily by the widespread adoption of AI tools: as we predicted in our previous report, these tools have played a major role in the discovery of vulnerabilities in third-party software. Meanwhile, these tools often contain security issues of their own. For example, OpenClaw, a popular AI project, ranked 12th among those with the highest number of vulnerabilities discovered and published in Q2, with over 200 CVEs registered during the reporting period. Finally, AI development tools are also contributing to the vulnerability landscape, since the quality of the code they produce can vary widely. Therefore, the rate at which new vulnerabilities are discovered will inevitably keep growing.

Total published vulnerabilities per month from 2022 through 2026 (download)

Next, we analyze the number of new critical vulnerabilities (CVSS > 9.0) over the same period.

Total critical vulnerabilities published per month from 2022 through 2026 (download)

As the chart shows, the number of published critical vulnerabilities jumped sharply in Q2. This is because using AI for vulnerability research makes it possible to analyze massive amounts of previously unexamined code, uncover new attack surfaces, and identify entire classes of vulnerabilities that have gone unnoticed for decades. In particular, AI was used to find a series of Dirty Frag vulnerabilities in the Linux kernel.

Exploitation statistics

This section presents statistics on vulnerability exploitation for Q2 2026. The data draws on open sources and our telemetry.

Windows and Linux vulnerability exploitation

Q2 2026 saw a new precedent in the publication of vulnerabilities in Windows components and exploits for these: researchers no longer waiting for CVE registration, let alone patches. A case in point: a researcher who goes by Nightmare Eclipse (also known as Chaotic Eclipse) published a list of new “named” vulnerabilities across various Windows subsystems. At the time the technical details were published, none of the vulnerabilities had been assigned a CVE identifier:

• BlueHammer: a local privilege escalation vulnerability in Windows Defender. During signature database updates, a time-of-check to time-of-use (TOCTOU) race condition occurs, allowing an attacker to substitute the directory where temporary update files are written. The researcher published a fully functional exploit for the vulnerability.
• RedSun: another logical vulnerability in Windows Defender with a working exploit. Suspicious and malicious files marked as “cloud” can be overwritten or restored to their original directory with elevated privileges. The exploit incorporates fragments of algorithms that make it possible to leverage various logical vulnerabilities in Windows, effectively combining a large number of popular exploitation techniques.
• YellowKey: a vulnerability that lets the user bypass BitLocker full-disk encryption and access system data through the Windows Recovery Environment (WinRE). A fully functional exploit was also published.
• GreenPlasma: a vulnerability that enables system object injection via the CTF loader for the Collaborative Translation Framework (CTFMON) service in Windows. The original publication included an exploit with limited functionality.
• RoguePlanet: yet another Windows Defender vulnerability that, like BlueHammer, stems from a TOCTOU issue, this time in the engine responsible for real-time system scanning. The published exploit uses the vulnerability to overwrite the system file wermgr.exe with a malicious one.
• UnDefend: another vulnerability in the Windows Defender service. This time, the exploit causes a denial of service and blocks updates.

Even though such cases remain isolated for now, we believe they’ll grow into a full-fledged trend. Early publication of exploits gives attackers an advantage over software developers, who are left with no time to fix the issues.

Veteran vulnerabilities in Windows software also remain relevant. These are the ones our solutions most frequently detect exploits for:

• CVE-2018-0802: a remote code execution (RCE) vulnerability in the Equation Editor component
• CVE-2017-11882: another RCE vulnerability also affecting Equation Editor
• CVE-2017-0199: a vulnerability in Microsoft Office and WordPad that allows an attacker to gain control over the system
• CVE-2023-38831: a vulnerability in WinRAR that involves improper handling of objects within an archive
• CVE-2025-6218 (formerly ZDI-CAN-27198): another WinRAR vulnerability allowing the specification of relative paths to extract files into arbitrary directories, potentially leading to malicious command execution
• CVE-2025-8088: a vulnerability similar in exploitation method to CVE-2025-6218. The attackers used NTFS Streams to circumvent controls on the directory into which files are being unpacked

The vulnerabilities listed here can be leveraged to gain initial access to a vulnerable system and for privilege escalation. This underscores the critical importance of timely software updates.

That said, the number of Windows users who encountered exploits declined slightly in Q2, hitting an 18-month low.

Dynamics of the number of Windows users encountering exploits, Q1 2025 – Q2 2026. The number of users who encountered exploits in Q1 2025 is taken as 100% (download)

Linux also hit a rough patch in Q2 2026. Specifically, the period saw the disclosure of the Dirty Frag family of vulnerabilities, which lets an attacker reliably escalate privileges within the operating system.

All the vulnerabilities published in Q2 2026 were, in one way or another, related to the Linux caching subsystem. Here are the ones being most actively exploited:

• CVE-2026-31431 (Copy Fail): a local privilege escalation vulnerability in the Linux kernel that lets an unprivileged user modify the page cache and gain root privileges. Especially dangerous for cloud and containerized environments
• CVE-2026-43284, CVE-2026-43500 (Dirty Frag): a family of vulnerabilities in the Linux networking subsystem (IPsec ESP and RxRPC) that lets a local user overwrite the page cache and escalate privileges to root
• CVE-2026-46300 (Fragnesia): a local privilege escalation vulnerability in the Linux kernel related to packet fragment handling and the page cache mechanism. It lets an unprivileged user gain root privileges and is also classified as part of the Dirty Frag family
• CVE-2026-31635 (DirtyDecrypt): a Linux kernel vulnerability that lets a local attacker escalate privileges due to improper handling of decryption operations and page cache data modification
• CVE-2026-43494 (PinTheft): a Linux kernel vulnerability that lets a local user gain elevated privileges due to errors in the memory page pinning mechanism
• CVE-2026-46331 (pedit COW): a vulnerability in the Linux kernel’s traffic control subsystem (tc-pedit) that exploits a flaw in copy-on-write to modify the page cache and subsequently escalate privileges to root

The vulnerabilities described above were quickly embraced by attackers. At the same time, our solutions continue to detect exploitation attempts targeting older vulnerabilities as well:

• CVE-2022-0847: a vulnerability known as Dirty Pipe, which enables privilege escalation and the hijacking of running applications
• CVE-2019-13272: a vulnerability caused by improper handling of privilege inheritance, which can be exploited to achieve privilege escalation
• CVE-2021-22555: a heap out-of-bounds write vulnerability in the Netfilter kernel subsystem
• CVE-2023-32233: another Netfilter subsystem vulnerability that allows for Use-After-Free conditions and privilege escalation through improper processing of network requests

Dynamics of the number of Linux users encountering exploits, Q1 2025 – Q2 2026. The number of users who encountered exploits in Q1 2025 is taken as 100% (download)

In Q2 2026, the number of Linux users who encountered exploits declined slightly compared to Q1. Given that a significant share of new vulnerabilities are tied to the operating system’s caching subsystem, we recommend installing patches as quickly as possible, or disabling vulnerable kernel modules if patching isn’t an option.

Most common published exploits

The distribution of published exploits by software type in Q2 2026 includes categories that haven’t appeared in the sample for a long time. For instance, we’re once again seeing exploits targeting SharePoint. It’s worth noting that while several vulnerability write-ups for Exchange and SharePoint were published during the quarter, most turned out to be fake, AI-generated research. While the articles and exploit source code themselves look fairly polished, they describe nonexistent problems in the software or its components — often close to genuinely vulnerable mechanisms — in order to mislead researchers. This type of attack is aimed at increasing the time it takes to detect real vulnerabilities. In some cases, the description of a nonexistent vulnerability came bundled with completely unrelated malware.

Distribution of published exploits by platform, Q1 2026 (download)

Distribution of published exploits by platform, Q2 2026 (download)

Vulnerability exploitation in APT attacks

We analyzed which vulnerabilities were exploited in APT attacks during Q2 2026. The rankings provided below include data based on our telemetry, research, and open sources.

TOP 10 vulnerabilities exploited in APT attacks, Q2 2026 (download)

In Q2 2026, a trend emerged in APT attacks toward exploiting new vulnerabilities right from the moment they’re published. As before, we’re also seeing a large number of zero-day vulnerabilities. The Langflow vulnerability deserves particular attention: it’s one of the first cases of an APT group exploiting AI technology, which many organizations are only just beginning to integrate. Because most of this tech is proprietary, it has a considerable number of security blind spots. Therefore, given the growing number of AI-based automation tools, we strongly recommend going beyond the usual patching and developing secure procedures for credential use and sensitive data handling in systems that rely on agents and LLMs.

C2 frameworks

In this section, we examine the most popular C2 frameworks used by APT groups and analyze the vulnerabilities targeted by the exploits that interacted with C2 agents in APT attacks.

The chart below shows the frequency of known C2 framework usage in attacks during Q2 2026, according to open sources.

TOP 10 C2 frameworks used by APTs to compromise user systems, Q2 2026 (download)

Sliver, Havoc, AdaptixC2, and Metasploit remain the most widely used C2 frameworks. After studying open sources and analyzing samples of malicious C2 agents that contained exploits, we determined that the following vulnerabilities were utilized in APT attacks involving the C2 frameworks mentioned above:

• CVE-2026-35273: a vulnerability in Oracle PeopleSoft PeopleTools that security vendors classify as server-side request forgery (SSRF). The details of the vulnerability have never been disclosed, although some research covers the post-exploitation steps
• CVE-2023-46604: an insecure deserialization vulnerability in Apache ActiveMQ that allows arbitrary code execution in the context of the service process
• CVE-2024-12356 and CVE-2026-1731: command injection vulnerabilities in BeyondTrust software that allow an attacker to send malicious commands even without system authentication
• CVE-2023-36884: a vulnerability in the Windows Search component that allows commands to be run on the system, bypassing the mark-of-the-web (MoTW) mechanism
• CVE-2025-53770: an insecure deserialization vulnerability in Microsoft SharePoint that allows for unauthenticated command execution on the server
• CVE-2025-8088 and CVE-2025-6218: similar directory traversal vulnerabilities in WinRAR that allow files to be extracted from an archive to a predetermined path, potentially without the archiving utility displaying any alerts to the user

These vulnerabilities show that attackers used them for initial access and privilege escalation on vulnerable systems, setting the stage for launching a C2 agent. They include both zero-day vulnerabilities and fairly well-known security issues.

LLM/AI tool vulnerabilities

This section analyzes data published in Kaspersky’s vulnerability knowledge base. We reviewed the Q2 2026 version of the knowledge base.

As mentioned above, AI tools, plugins, and technologies have proven fairly effective at automating the search for problematic code and anomalous behavior. The high speed at which new vulnerabilities are being discovered has naturally created a need to fix them just as quickly. AI is often used for this too, which increases the volume of code being generated. However, neither code written without human involvement nor AI-generated advice is always correct.

The chart below covers registered vulnerabilities in AI tools for 2025–2026.

Number of published vulnerabilities in LLMs, AI tools, and plugins with similar functionality, 2025–2026 (download)

As the charts show, AI tools are racking up a substantial number of registered vulnerabilities, and that number keeps growing quarter over quarter. It’s also worth looking at how AI tool vulnerabilities break down by type, according to the CWE system:

TOP 6 vulnerability types in products that implement or use AI/LLM logic, 2025–2026

Interestingly, vulnerabilities of an undetermined type have ranked first in every quarter since the start of 2025. Traditionally-made software has the same issue, and it doesn’t look like the growing number of AI tools will fix it. It’s also notable that the list includes classes CWE developers themselves don’t recommend using for vulnerability classification, since they lump together a whole range of more specific types. CWE-284 is an example of this.

Looking at the most common classes, the key issues found in AI-related software can be summed up as follows:

• Inadequate access control over critical system objects
• Improper implementation of authentication and authorization mechanisms
• Injections

It’s worth noting that injection-related vulnerabilities were relatively rare before AI agents took off (previously, they mostly affected web apps). Recently, though, these security issues have become relevant again.

Looking back at a year and a half of the AI boom, one conclusion stands out regarding registered vulnerabilities: AI tool developers are more focused on expanding functionality than on security. This is worth keeping in mind when using these tools. Let’s look at the projects and applications that either integrated AI tools or offered them as the core product. Below is a list of the those with the highest number of registered vulnerabilities for 2025–2026.

TOP AI/LLM-related projects by number of published vulnerabilities, 2025–2026 (download)

Notable vulnerabilities

This section highlights the most significant vulnerabilities published in Q2 2026 that have publicly available descriptions. Since the above already covers several significant vulnerabilities published during the reporting period, this section consists mainly of LLM/AI tool vulnerabilities.

CVE-2026-25253: a gatewayUrl vulnerability in OpenClaw

The issue stems from the fact that the OpenClaw user interface trusts the value of the gatewayUrl parameter passed in the URL and automatically establishes a WebSocket connection to the specified address. During this connection process, it sends an authentication token without any additional user confirmation.

The attack algorithm exploiting this vulnerability works as follows:

1. The application obtains a critical connection address from an external source (the gatewayUrl URL parameter), which is controlled by the attacker.
2. There is no validation before use.
3. The client automatically initiates a connection to the address specified in the parameter, which belongs to the attacker.
4. While connected, the application sends credentials (an access token) to the specified address.

If the attacker obtains a valid token, the consequences depend on that token’s level of access within the system. In general, this could lead to:

• User session compromise
• Execution of operations on the user’s behalf
• Modification of the AI agent configuration
• Unauthorized access to tools and resources connected to the agent
• Under certain OpenClaw configurations, further compromise of the host running the agent

It’s worth noting that the risk of exploitation arises from a combination of several factors: the automatic connection and token transmission, the lack of address trust verification, and the high privileges granted to the local AI agent.

CVE-2026-41948: a path traversal vulnerability in the Dify AI platform

The vulnerability lets an authenticated user craft a request that enables the application to escape its permitted tenant and gain access to internal REST APIs that weren’t meant for that user. The root cause is insufficient normalization and validation of the URL path before it’s passed to the internal service.

Depending on the Dify configuration, the consequences can include:

• Unauthorized access to internal service interfaces
• Breach of isolation between workspaces
• Exposure of internal service information
• Conditions favorable to further attacks when combined with other vulnerabilities

The use of Dify in enterprise AI platforms is particularly risky, since internal services there tend to hold elevated privileges.

CVE-2026-45386: an improper access control vulnerability in Open WebUI

In Open WebUI, pin/unpin operations on messages are write operations, since they modify that message’s metadata (is_pinned, pinned_by, pinned_at). In vulnerable versions, however, before performing these actions, the API only checked for read access to the channel (a chat between a user or group and the AI) containing the message, not permission to modify its content. As a result, a user with a role limited to viewing messages could still change a message’s pinned status.

The vulnerability’s mechanism works as follows:

1. The user initiates an action that changes the state of an object.
2. The application treats this action as a regular read request.
3. Only channel view permission is checked.
4. The application performs a write without verifying the required user authorization.

This violates one of the fundamental principles of access control models — namely, that any operation that changes the state of data must be checked for the appropriate write or moderation permissions, regardless of whether the object itself is readable.

Although the vulnerability doesn’t lead to arbitrary code execution or compromise of sensitive data, it can affect data integrity and collaborative workflows. Potential consequences of exploitation include unauthorized pinning or unpinning of messages, disruption of channel moderators’ and administrators’ activities, changes to the display order of important information, and even the potential spread of false or misleading information by altering the channel containing a pinned message.

Open WebUI is widely used as an interface for interacting with local and enterprise LLMs. In these systems, pinned messages often contain important instructions, announcements, or tips for users. The ability to modify them with minimal privileges can disrupt collaborative workflows, cause confusion, and undermine trust in information published by administrators and moderators.

CVE-2026-45501: a vulnerability in Microsoft Exchange

The vulnerability stems from improper neutralization of user input when generating Exchange web pages. As a result, the browser may interpret specially crafted data as active content instead of plain text.

Although Microsoft categorizes the potential impact of exploiting this vulnerability as spoofing, flaws like this can lead to alteration of displayed content, imitation of trusted interfaces, actions on behalf of the user within an active session, and abuse of user trust.

It’s worth noting that issues like this are still relevant in modern software, given that mechanisms like Content Security Policy and various parsers were specifically created to help developers neutralize dangerous parts of user page content.

Conclusion and advice

Q2 brought the first significant results of AI automation adoption in software development and vulnerability hunting tools. This research shows that beyond traditional patch management, organizations now need real-time monitoring of systems and access controls, since infrastructure and everyday applications now contain far more AI functionality that could lead to compromise.

Accordingly, besides quickly detecting infrastructure vulnerabilities and managing security patches, modern enterprise-grade security solutions need to provide a broad range of preventive measures for tracking the overall health of systems and workstations. Kaspersky Next meets these requirements by combining proactive mechanisms with the ability to respond promptly to emerging threats.

In May 2026, we discovered a new cyber-espionage campaign by the Armored Likho group, also known as Eagle Werewolf, that targets private individuals and organizations across various industries in Russia, including major corporations, the public sector, IT, and education. The attackers used a fake app as bait that mimics a service for donations. However, the most interesting part of this campaign isn’t the initial infection method – it’s the malicious implants the attackers use for cyber-espionage.

We’ve written previously about recent Armored Likho attacks, but our analysis shows that the campaign discussed below has more in common with the group’s activity from February. That said, the attackers have significantly expanded their arsenal.

During our research, we found a new cyber-espionage toolkit written in Rust: the Still Toolkit. One of its components, Still Sync, steals Telegram session data to gain ongoing access to the victim’s account. With this stolen data, attackers can leverage the Telegram API to automatically pull chat logs, media files, and other information from the account.

The second component, Still Audio, is an implant for covert audio surveillance. It analyzes the incoming audio stream, automatically detects speech, records conversations, and sends the recordings to a command-and-control server.

In this article, we’ll look at the initial infection method, how the new Still Toolkit components are built, and the technical details of how they operate.

Kaspersky products detect this threat as Trojan.Win64.Agent.* and HEUR:Backdoor.Win32.Generic.

Background

Armored Likho’s malicious activity has been documented several times before: in November 2024, and in February and July 2026. The current campaign shows significant overlap with the November and February campaigns, which used malicious droppers disguised as documents and applications related to Starlink activation or fundraising efforts as the initial infection vector. This campaign also uses fundraising as its lure. At the same time, our research uncovered a number of new tools that point to the attackers expanding their capabilities.

Initial infection

The infection chain starts with an app that mimics a donation service. As of this writing, the app distribution method remains unknown. During our research, however, we obtained several samples posing as apps from different Russian foundations.

In reality, the app is a dropper. Its developers wrote it in Rust on top of the popular Tauri framework, and it has a graphical interface designed to deceive the user. After launch, it displays a login form that asks for a password, presumably one the attackers supplied.

The login form

After the user enters a valid password, they see a catalog of donatable items. The app pulls item and category information from orderapiserver[.]info through the public/categories and public/products endpoints. A clickable catalog makes the app look legitimate. While the user browses the items, the dropper quietly decrypts and launches the payload for the next stage in the background.

Our analysis shows that the mechanism for decrypting the payload and launching subsequent stages hasn’t changed since the February campaign. However, we found a new cyber-espionage toolkit – the Still Toolkit – made up of two components: Still Sync and Still Audio.

Still Sync

Still Sync is a stealer written in Rust that steals Telegram session data. However, its capabilities don’t stop there. With this stolen data, Sync can log in to the victim’s account and pull messages and media files through the Telegram API.

Architecturally, Sync is an asynchronous application based on the Tokio library. It talks to the server over gRPC and serializes messages with FlatBuffers. It supports both HTTP and HTTPS as transport protocols; the URL of the command-and-control server determines which one it uses.

How it works

When Sync launches, the attackers set several environment variables. Before starting any malicious activity, the implant pulls configuration parameters from these:

• STILL_SYNC_ADDR: the address of the command-and-control server. By default, this is https://tg4service[.]com:443.
• STILL_SEND_PATH: the path to the tdata
• STILL_TELEGRAM_PASSCODE: the password for decrypting the tdata folder, if Telegram data encryption is enabled on the victim’s device.

Sync also supports several command-line arguments:

• --console: runs as a console application. If this parameter is absent, the implant creates a TReload service to keep running in the background.
• --version: prints version information and exits.
• --firefly: launches a trace thread that monitors the program’s operation. It writes error messages to a hidden file, bin, located in the same folder as the main executable.
• --db: turns on debug mode with detailed logging.

Example Still Sync logs

Once it launches, the malware begins registering the device with the C2 server. To do this, Sync collects the following information about the victim’s system:

• Motherboard serial number
• CPU ID
• System UUID
• BIOS serial number
• Computer domain name

The malware combines the collected data into a single string with a colon as the separator. It then hashes that string with SHA-256 and stores the resulting hash under the key sysmarker. Worth noting: other Armored Likho tools, AquilaRAT included, use this same hashing algorithm.

Sync then serializes a package containing all the collected information and the agent version, and sends it in a POST request to /still.rpc.Sync/RegisterMachine. The response contains a machine_id value, which Sync uses to identify itself in subsequent requests.

Once registration succeeds, Sync sends a POST request with the machine_id parameter to /still.rpc.Sync/GetMachineSettings. The server responds with the following settings:

• enabled: triggers malicious activity on the infected device.
• scan_portable: turns on extended scanning when searching for the tdata We’ll cover this feature in more detail below.
• fetch_telegram: if this parameter is on, Sync attempts to log in to Telegram and extract data. We’ll cover this feature in more detail below.
• download_channels: if this parameter is off, Sync skips channel dialogs when exfiltrating Telegram data.

These parameters have no default values, so Sync doesn’t perform any malicious actions until the registration and settings-retrieval processes both complete successfully.

Telegram data collection

Before stealing a Telegram session, Sync searches for the tdata folder, unless the STILL_SEND_PATH variable is already set. The list of search paths includes both standard and nonstandard directories, if the scan_portable option is turned on:

• C:\Users\<username>\AppData\Roaming\Telegram Desktop\: the standard Telegram Desktop installation directory.
• C:\Users\<username>\AppData\Local\Packages\<package_folder>\LocalCache\Roaming\: the installation directory for the Microsoft Store version. Sync identifies the package folder by a name that contains the string TelegramMessenge.
• C:\: used for the extended search (if the scan_portable option is on).

Sync then sends a POST request with a list of files from the tdata folder to the /still.rpc.Sync/CheckFiles endpoint. The server responds with the following values:

• snapshot_id: an identifier the server assigns to the current data snapshot.
• present: a list of file paths that are already present on the server.

This lets the C2 server avoid re-receiving files it already has. In addition, if Sync can’t access files on disk through standard methods, it falls back on three mechanisms that abuse the SeBackupPrivilege privilege:

• Opening files with the CreateFileW function using the FILE_FLAG_BACKUP_SEMANTICS parameter
• Creating a backup copy through the Shadow Copy service and reading files from there
• If the previous methods all fail, attempting to copy the file using the Robocopy utility in backup mode

Beyond stealing Telegram session data, Sync can carry out full-scale collection of user information from the messaging app. When the fetch_telegram option is on, it launches a separate thread that authenticates to the chat app using the previously obtained tdata. Once authentication succeeds, Sync gains access to the account data and sends the following collected information to the server:

• User details, such as username, phone number, first and last name
• Information about private chats, groups, or channels, such as chat name and ID, the member list, and so on
• Dialogs from private chats, groups, and channels (if the download_channels option is on)
• Media files under 250MB: photos, documents, stickers, and contacts

Still Audio

Still Audio is an audio surveillance implant written in Rust. Its main job is to analyze the incoming audio stream and start recording voice when certain conditions are met – we’ll cover those in the next section. Architecturally, Still Audio largely mirrors Sync and uses the same mechanisms for communicating with the C2 server.

On launch, Still Audio performs a sequence of actions:

• It extracts libmp3lame.dll, a file stored inside the executable. This is a library used to encode audio data.
• If the --console command-line argument is absent, the implant creates a service named auxhost, connects to it, and continues running in the background.
• While running in the background, it creates a file, logfile.log, to write logs to.

Next, Still Audio retrieves the C2 server address. As with Sync, it stores the URL in an environment variable – in this case, STILL_AUDIO_SYNC_ADDR. If that variable isn’t set, it falls back to STILL_SYNC_ADDR, which shows the two modules are compatible with each other. If neither variable is set, it uses the default URL, https://srwinservice[.]com.

Still Audio also uses the Dead Drop Resolver technique as a fallback mechanism for obtaining the C2 address. If the current server stays unreachable for three days, the tool tries to pull the current C2 URL from a GitHub repository. In the sample under analysis, we found the following URL for the page containing C2 information: hxxps://raw.githubusercontent[.]com/mmarln/pi-mono/refs/heads/main/packages/pods/src/array12.json

Encrypted C2 address inside the GitHub repository

The repository, a fork of a popular project, contains the server URL Base64-encoded and encrypted with the Blowfish algorithm in ECB mode, using the key 5c8e153228edd3c6cbf75684 (lowercase string). Older AquilaRAT samples use this exact same algorithm and key.

Once it obtains the current C2 address, the Audio module starts a registration process similar to Sync’s, but through a different endpoint:

/still.rpc.Audio/RegisterAudioMachine. Also, unlike Sync, Audio sends a list of available audio input devices along with the system information.

The server responds with settings for the implant:

• machine_id: a unique identifier for the current device.
• vad_threshold: the threshold value for the VAD (Voice Activity Detection) algorithm. Expressed as a decimal fraction, it represents a proportion of the maximum sound level the input device can pick up. Sound above this threshold counts as voice activity. The default vad_threshold is 02.
• max_silence_duration: the number of audio samples with a VAD value below the set threshold after which the implant considers the recording finished.
• max_buffer_size: the maximum buffer size for recorded audio data.
• active_device: the name of the input device selected for recording, from the list of available devices.

The eavesdropping process

Still Audio works with raw audio samples it captures directly from the input device. To detect voice activity, it implements an algorithm based on Root Mean Square (RMS), a lightweight signal-processing method that distinguishes speech from silence by measuring the audio signal’s average power over time. The implant doesn’t rely on any third-party libraries here; it implements all the calculations itself.

The implant compares the calculated RMS value against the vad_threshold parameter. If RMS meets or exceeds this threshold, recording starts. To avoid losing the beginning of the recording, Still Audio uses a pre-buffer, a size-limited buffer that stores samples from just before the current recording moment. A sequence of max_silence_duration samples (320 by default) with RMS values below the threshold signals the end of the recording. For example, with a standard headset running at a 44.1kHz sampling rate, recording stops after roughly 7ms of silence.

Interestingly, the Audio module makes no attempt to hide its use of the microphone: its name shows up in Windows settings. In the sample we examined, the file was saved to disk as IntAudio.exe, and it appeared in the list of apps using the microphone as “Intel Audio”:

The malicious module in the list of apps using the microphone

Before sending recordings to the server, the implant uses the libmp3lame library to encode the raw audio samples. It sends the recording files via a POST request to /tgfrg, adding a Client-Id header containing the machine_id obtained during registration to identify the device.

Infrastructure

This campaign draws on a broad set of hosting providers and domains registered at different points in time, which suggests the attackers are trying to make their infrastructure harder to detect. We found no direct overlap in domains or IP addresses with the February campaign. Even so, the two infrastructures share some similarities:

• They use the same hosting providers, with the ASNs 149440, 202448, and 215311.
• Their domain names follow similar naming patterns that mimic Windows system services and update mechanisms.

Victims

In this campaign, we’ve determined that the attackers’ primary targets are users in Russia. Most victims are private individuals, though the corporate sector, government organizations, IT companies, and educational institutions are also affected.

Attribution

This campaign has been using both new tools and malware families documented in BI.ZONE’s February report. While some components turned up for the first time, they show significant code-level overlap with malicious tools seen in earlier Armored Likho campaigns. Based on these overlaps, along with additional technical artifacts, we’re highly confident the Armored Likho group is behind the campaign. The overlaps we identified include:

• Identical dropper architecture in the February and current campaigns, which includes the use of the Tauri library to build the graphical interface, a similar user-input handler, a payload with the ICRYPTMP header, and the same multi-part encryption format.
• The same encryption algorithm and key used in AquilaRAT from the previous campaign and in the Still Audio module from the current campaign, both implementing the Dead Drop Resolver technique.
• Identical logic for generating the sysmarker value in older AquilaRAT samples and in the Still toolkit from the current campaign. The algorithms match down to the PowerShell commands used to collect system information.
• Substantial infrastructure overlap, which includes the hosting providers and domain-naming patterns described in the Infrastructure section.

Takeaways

The campaign described in this post shows Armored Likho’s toolkit evolving, with the group steadily expanding its cyber-espionage capabilities. Beyond the components we already knew about, the attackers rolled out new modules that let them not only access Telegram data but also conduct audio surveillance on victims. Together, these capabilities significantly widen the range of information attackers can collect in a single compromise.

One point deserves particular attention: the new tools form a cohesive set, sharing similar architecture, C2 communication mechanisms, and common implementation elements. This points to the group building out its own tool ecosystem, designed for long-term use and further expansion.

The emergence of new, specialized modules shows the attackers aren’t just trying to preserve their existing capabilities – they’re working to make intelligence-gathering more effective by controlling multiple communication channels at once.

Indicators of compromise

Additional information about this threat, indicators of compromise included, is available to customers of Kaspersky Threat Intelligence Reporting. Contact intelreports@kaspersky.com for more details.

File hashes
Droppers
C1D1EE16B92E6A138FFA048855F75D7D
17674B250D8B422A50A86C9FF207186D
62801F6223E860A7CCA271522E303B2D

Still Sync
68F0365D2FA8C828D012D8859E52A773
4BD7C352AE277B0E38D07BEEDD4DD507
D4BC09FB10EA2A5DC0BCBEEDA5E5AFDD

Still Audio
2CA8ADBAB98EBE305EACF272CF48F5A0
3AC41B097236A7723821848AE31EF141
439255736797BC88BD19F282449E0436

Domains
orderapiserver[.]info
tg4service[.]com
srwinservice[.]com
screenserv[.]com
windowserv[.]net
managementapiservice[.]com
service8date[.]com
updateservs[.]com