Cyware Daily Threat Intelligence, May 12, 2025

Daily Threat Briefing • May 12, 2025

Not every AI tool on Facebook is what it claims to be. Threat actors are luring users with fake content-generation apps, pushing Noodlophile malware to tens of thousands of victims. Disguised as AI-powered services, these tools trick users into executing malicious files that steal browser credentials, crypto wallet data, and other sensitive information through a layered infection chain.

OtterCookie is getting smarter with every version. The North Korea-linked group WaterPlum is actively using the malware to breach financial institutions and crypto platforms worldwide. Since late 2024, new iterations have added capabilities like credential theft from browsers and file uploads - each version revealing signs of modular development and distinct authorship.

A flaw in Linux’s nftables is turning dirty sets into dangerous ones. A critical double-free vulnerability allows local attackers to escalate privileges by exploiting memory corruption. Kernel-level fixes are already rolling out to improve cleanup routines and patch the vulnerable destroy path.

Top Malware Reported in the Last 24 Hours

Fake AI tools deploy Noodlophile malware

Threat actors are leveraging fake AI tools on Facebook to distribute the Noodlophile malware, targeting over 62,000 users. These tools masquerade as legitimate AI-powered content creation services, tricking users into downloading malicious files. The malware harvests sensitive data, including browser credentials and cryptocurrency information, through a complex infection chain initiated by a deceptive executable.

Contagious Interview drops OtterCookie

WaterPlum, a North Korea-linked attack group, has been using the OtterCookie malware to target financial institutions and cryptocurrency operators globally. Since its introduction in September 2024, OtterCookie has evolved through multiple versions, with v3 featuring file upload capabilities and Windows support, while v4 adds stealer modules for credentials from Google Chrome, MetaMask, and Brave browser. Differences in coding suggest varying developers for these modules. The group’s activities have been notably observed in Japan, with ongoing updates to the malware.

Malicious npm packages abuse Telegram

Socket has uncovered two malicious npm packages, ‘pumptoolforvolumeandcomment’ and ‘debugdogs’, designed to steal cryptocurrency wallet keys and sensitive data from users of the BullX trading platform. The malware targets Linux and macOS systems, searching for Base58-encoded keys and files related to BullX. It employs obfuscated payloads and uses a Telegram bot for real-time exfiltration of stolen data. The secondary package, debugdogs, acts as a wrapper to facilitate the spread of the primary malware.

Top Vulnerabilities Reported in the Last 24 Hours

Critical bug in Linux nftables

A critical double-free vulnerability (CVE-2024-26809) has been identified in the Linux kernel's nftables subsystem, allowing local attackers to escalate privileges and execute arbitrary code. This flaw occurs in the `nft_pipapo_destroy()` function when a "dirty" set leads to improper handling of elements, causing memory corruption. Exploitation involves creating a pipapo set, marking it as dirty, and deleting it, which triggers the vulnerability. Fixes include updates to the Linux kernel, such as improving the cloning process, integrating nft_set_pipapo into the commit protocol, and ensuring proper element release from the destroy path.

Microsoft patches critical vulnerabilities

Microsoft has identified four critical vulnerabilities in Azure and Power Apps. CVE-2025-29813 is a severe Azure DevOps flaw with a CVSS score of 10.0, allowing attackers to manipulate pipeline job tokens. CVE-2025-29827, affecting Azure Automation, scores 9.9 and enables privilege escalation for authenticated users. CVE-2025-29972 is a server-side request forgery (SSRF) vulnerability in Azure Storage, also scoring 9.9, permitting spoofing attacks. Lastly, CVE-2025-47733, an SSRF vulnerability in Power Apps, scores 9.1 and could allow unauthorized information disclosure.

Related Threat Briefings

May 9, 2025

Cyware Daily Threat Intelligence, May 09, 2025

Old routers are becoming cybercrime goldmines. The FBI has warned that end-of-life routers are being hijacked with malware like TheMoon and sold on proxy networks such as 5Socks and Anyproxy. These compromised devices are used for crypto theft, cybercrime-as-a-service, and even espionage. Meanwhile, Chaya_004 is exploiting SAP systems on a global scale. The Chinese threat group was found exploiting CVE-2025-31324, a critical flaw in SAP NetWeaver, to gain remote code execution. Their campaign has hit hundreds of organizations globally, leveraging a Golang-based shell named SuperShell and tools hosted on Chinese cloud platforms in a coordinated effort. Phishing pages are climbing the search results and duping users. The FreeDrain campaign is using SEO poisoning and free-tier platforms like GitBook and Webflow to host fake wallet interfaces that harvest seed phrases. Over 38,000 phishing subdomains have been linked to this large-scale crypto scam, believed to originate from India and powered by generative AI. For detailed Cyber Threat Intel, click ‘Read More’.

May 8, 2025

Cyware Daily Threat Intelligence, May 08, 2025

COLDRIVER’s latest malware, LOSTKEYS, is now in play. The Russian state-backed group is deploying this tool to steal files and system data from advisors, journalists, NGOs, and individuals linked to Ukraine. Delivered via ClickFix, LOSTKEYS expands COLDRIVER’s toolkit beyond credential phishing, signaling a continued focus on espionage and document theft. SysAid users, it’s time to patch. Multiple vulnerabilities—ranging from XXE to OS command injection—have been fixed in version 24.4.60. Researchers uncovered the flaws and even released a PoC, warning that 77 unpatched instances remain exposed online. Given SysAid’s large user base, the window for exploitation remains a real concern. A clever lure is fueling a quiet takeover in Brazil. Cisco Talos uncovered a campaign abusing the country’s electronic invoicing system to trick victims into installing remote monitoring tools. The attackers use free trials of RMM products to hijack machines, often targeting high-level accounts in finance, education, and government.

May 7, 2025

Cyware Daily Threat Intelligence, May 07, 2025

Agenda’s playbook just got upgraded. The ransomware group has added two new tools: SmokeLoader and a stealthy .NET-based loader called NETXLOADER. The latter leverages techniques like JIT hooking and AES decryption to deploy ransomware and loaders directly into memory, while SmokeLoader injects payloads into explorer.exe using anti-analysis tricks. In a different vein, a bogus debugging package was found duping Discord bot developers. The malicious PyPI package ‘discordpydebug,’ posing as a helpful tool, is designed to steal data and execute commands. Downloaded over 11,500 times, it exploited community trust, mirrored by over 45 npm lookalikes in what appears to be a broader campaign. AWS patched a critical bug in Amplify Studio version 2.20.3. CVE-2025-4318 allowed attackers to inject JavaScript into UI components during render time via the amplify-codegen-ui package. No active exploitation was detected.

May 6, 2025

Cyware Daily Threat Intelligence, May 06, 2025

Mamona isn’t calling home but it’s still locking you out. This offline ransomware strain skips data theft and C2 communication altogether, instead encrypting files using its own cryptographic routines. Though it avoids detection through network monitoring, a working decryption tool is already available for victims. Google’s latest Android update delivers fixes for around 50 vulnerabilities, including a critical zero-day in FreeType, which was already under active attack. The patch batch covers issues across Android OS, MediaTek, Qualcomm, Wear OS, and Automotive OS, reinforcing security across a wide device ecosystem. CoGUI is fueling a phishing surge in Japan by spoofing trusted brands like Rakuten and PayPay. The kit uses geofencing and browser fingerprinting to narrow its targets, directing victims to convincingly faked login pages that siphon credentials and payment details with precision.

May 5, 2025

Cyware Daily Threat Intelligence, May 05, 2025

Corporate HR teams are the latest target in a spear-phishing spree by Venom Spider. Disguised as job applications, these emails deliver More_eggs backdoor, now upgraded with advanced features. The attack leverages advanced evasion tactics to slip past defenses and establish persistent access. A fresh batch of vulnerabilities from MediaTek could leave millions of devices open to attack. The most severe enables remote DoS via rogue base stations. Other flaws impact encryption, memory handling, and certificate validation across phones, tablets, and smart devices running Android 13 to 15. Luna Moth is back, and it's using helpdesk lookalikes to worm its way into corporate inboxes. The campaign spoofs trusted brands with phishing emails and fake domains, siphoning credentials and financial data. So far, 50 domains tied to this operation have been linked to breaches in legal, healthcare, and finance sectors.

May 2, 2025

Cyware Daily Threat Intelligence, May 02, 2025

In the shadows of the cybersecurity landscape, MintsLoader emerges as a formidable adversary, orchestrating a multi-faceted infection strategy that deploys the notorious GhostWeaver RAT. This malware loader ingeniously employs a blend of obfuscated JavaScript and PowerShell scripts to execute its malicious agenda, utilizing sophisticated evasion techniques and a DGA for C2 communications. Two new malware families—TerraStealerV2 and TerraLogger—are now linked to the Golden Chickens threat group. TerraStealerV2 steals credentials and crypto wallet data but can’t yet bypass Chrome’s built-in encryption, suggesting it’s still in progress. TerraLogger, meanwhile, operates as a standalone keylogger with no exfiltration module, hinting at a plug-and-play role in a broader MaaS toolkit. A fake refund for a real blackout. A phishing campaign is spoofing TAP Air Portugal, preying on confusion after the Iberian power outage. Victims are promised compensation for delayed flights, only to be directed to phishing pages that steal payment details. The lure is simple, but effective, and the infrastructure, as usual, runs through compromised WordPress sites.

May 1, 2025

Cyware Daily Threat Intelligence, May 01, 2025

Some PyPI packages are doing more than importing functions. Researchers uncovered seven malicious Python packages under the “Coffin” naming scheme, using Gmail’s SMTP service as a stealthy C2 channel. Hardcoded credentials let the malware establish outbound tunnels via port 465, send activation signals, and quietly exfiltrate data. Apache ActiveMQ users should be on high alert. A flaw in its .NET Message Service library now lets remote attackers inject code by abusing deserialization. All versions prior to the latest update are at risk, so prompt patching is essential to protect your systems. Emails that look like official Social Security alerts are turning into dangerous gateways. In the new Molatori phishing campaign, attackers send messages claiming your Social Security Statement is ready to download. Once you install the ScreenConnect client they push, remote access is granted, and sensitive data gets swept away, thanks to convincing links from compromised WordPress sites.

Apr 30, 2025

Cyware Daily Threat Intelligence, April 30, 2025

A familiar threat actor is back with sharper tools. A new spear-phishing wave from Earth Kasha is hitting government targets in Taiwan and Japan. Using a malicious Excel file, the group drops an updated ANEL backdoor and maintains stealthy persistence with SharpHide. The second-stage malware hides C2 traffic behind DNS over HTTPS - signaling renewed focus on quiet, long-term access. TheWizards aren’t casting spells, they’re hijacking software updates. This Chinese-linked threat group is abusing IPv6 SLAAC to perform AitM attacks with the Spellbinder lateral movement tool. Spellbinder redirects traffic to attacker-controlled domains to deliver malware-laced updates to apps like Tencent QQ and Sogou Pinyin. Once inside, a modular backdoor takes over. New vulnerabilities in Apple’s AirPlay protocol, collectively dubbed AirBorne, expose billions of devices to remote code execution without user interaction. Attackers could silently hijack devices via use-after-free and buffer overflow vulnerabilities, affecting not only Apple hardware but also third-party products using the AirPlay SDK.

Apr 29, 2025

Cyware Daily Threat Intelligence, April 29, 2025

Sharp and TX stealers are back, donning a new cloak - named Hannibal Stealer. It is going after credentials from browsers, crypto wallets, FTP clients, and VPN apps. It even captures Discord tokens and Steam sessions. While its core is largely unoriginal, its control panel and delivery updates point to a more commercialized approach, complete with dark web subscriptions. Privilege escalation is just a pointer error away. Attack of the Vsock exposes a critical use-after-free bug in the Linux kernel’s VMware vsock driver. A mishandled reference count allows attackers to reclaim memory and more. With exploit code already circulating, this one has the potential to hit millions of virtualized systems. They sound urgent. They look official. And they’re emptying wallets. TA2900, a newly identified BEC actor, is impersonating landlords to trick renters in France and Canada into sending payments to attacker-controlled accounts. Messages come from compromised university inboxes and often swap out IBANs mid-thread.

Apr 28, 2025

Cyware Daily Threat Intelligence, April 28, 2025

Ransomware groups aren’t always the ones breaking the door open. Researchers have uncovered ToyMaker, an initial access broker selling network entry to ransomware groups. Using a custom malware strain called LAGTOY, ToyMaker establishes reverse shells and executes commands on compromised systems. After gaining access, ToyMaker hands over the keys - leaving ransomware operators to finish the job. Industrial remote access just got a lot more dangerous. Three vulnerabilities have been discovered in the IXON VPN client, affecting both Windows and Linux. Attackers can exploit temporary file handling flaws to escalate privileges. Full details remain under wraps until patches are ready. One bad notification and your iPhone could be stuck in an endless reboot loop. A critical iOS flaw allowed malicious apps to abuse the Darwin notification system and brick devices with a single command. Apple’s fix in iOS 18.3 now requires special entitlements to send sensitive notifications but unpatched devices remain at risk.

Apr 25, 2025

Cyware Daily Threat Intelligence, April 25, 2025

An APT group with deep roots in Southeast Asia is quietly siphoning data through everyday cloud platforms. Earth Kurma has been active since late 2020, targeting government and telecom entities across the Philippines, Vietnam, Thailand, and Malaysia. Its extensive toolkit enables credential theft, stealthy surveillance, and Dropbox-based exfiltration. One patch, two problems. Microsoft’s fix for CVE-2025–21204 may have closed a privilege escalation bug but it opened the door to a DoS vulnerability. By creating junction points, non-admin users can block future Windows updates from installing. There’s no fix yet, and unless the rogue junction is manually removed, security updates will fail silently. Logos, lures, and lies - the Power Parasites campaign is using all three. Fraudsters are impersonating global energy brands to scam victims in Bangladesh, Nepal, and India through fake jobs and investment offers. With over 150 domains and active Telegram and YouTube promotion, the scheme blends fake onboarding forms with identity theft and financial fraud, all under the guise of renewable energy.

Apr 24, 2025

Cyware Daily Threat Intelligence, April 24, 2025

It starts with a fake sales order and ends with FormBook silently stealing your data. A recent phishing campaign has been abusing a long-patched Microsoft flaw to deliver a fileless variant of the malware. The attack uses a Word document, a fake PNG file follows, carrying an encrypted payload that’s decrypted and run entirely in memory - leaving no trace on disk. A single request is all it takes to bring SonicWall firewalls to a halt. CVE-2025-32818 targets the SSLVPN Virtual Office interface and can trigger a DoS condition on affected Gen7 appliances. The vulnerability, rated 7.5 in severity, has been patched in recent firmware updates. Signal and WhatsApp are the new frontline for cloud compromise. Russian actors are running OAuth phishing campaigns against Microsoft 365 users tied to Ukraine and human rights work. Victims are tricked into handing over legitimate Microsoft auth codes, often via chats impersonating European officials.