Talos Content Manager Amy introduces themself, shares her unconventional journey into cybersecurity and reports on threats masquerading as AI installers.

Thursday, May 29, 2025 14:00

Welcome to this week’s edition of the Threat Source newsletter. 

In the words of Game Changer host Sam Reich, “And your host, me! I’ve been here the whole time!”  

Okay, maybe it’s not the whole time, but for the past three months, I've been settling into my role here at Cisco Talos. Editing blogs, writing and publishing social media posts, and organizing this newsletter every week — I've been working behind the scenes to ensure everything runs smoothly and delivers the most helpful information to the cybersecurity community. 

I often get raised eyebrows when I mention that, prior to my last job as a technical writer, I had never worked in STEM. I don't blame them, because how could someone who had never opened Terminal (and admittedly, up until last month sometimes forgot what it was called) end up with a job offer from Talos? 

My college degree is in anthropology, or the study of humans and culture, past and present. Though my niche research interest was/is Malaysian culture, LGBTQ+ history, and politics (even getting a research grant to travel to peninsular Malaysia for a month), my first career out of college was fundraising for a homeless services nonprofit in Arlington, Virginia. After I moved to another state, I held a content writing position at a startup, where I wrote fundraising letters and emails for a portfolio of over 200 nonprofits.

Learning the four-string Malaysian sape’. 

While I felt invested in these organizations' missions, I began to feel understimulated. I craved a career that would build on my experiences and skills while giving me the chance to learn and grow in new, exciting ways. While searching for new jobs on LinkedIn, I happened upon a nearby physical layer encryption startup that was seeking a technical writer. I had no clue what the physical layer even was, so I was grateful when they took a chance on hiring me. I found that my background in anthropological research, as well as my ability to adapt content for a lot of different audiences, became a huge asset in technical writing. 

I’ve always said that if I could magically be paid to go to school forever, I would. Technical writing (and its cousins, like my current position) is as close as I can get! After I joined Talos, I found that people here are incredibly kind and very patient. Like Jon Munshaw, the person who held this role before me, my favorite question to Talos researchers is “Can you explain this to me like I’m your grandmother?” Not only does it help me grasp the concepts they're sharing, but it also helps me find the clearest way to communicate them. 

Talosians are brilliant people, and I’m only human, so it’s easy to feel like you don't belong when you don’t have a STEM background. In a recent moment of doubt, I remembered that Joe had published a newsletter introduction about imposter syndrome two days after I started at Talos. One line stuck out to me: “You are where you are because others saw value in your work.” 

As I took in the sentence, I realized that it was entirely true. If there's one thing that I’ve learned over the past few months, it’s that everyone you meet has something to teach and everyone has something to learn. Our collective knowledge and experience are gifts we share with one another. I hope that the content I edit and produce will bring value to you. 

So what kind of content will I bring to this newsletter? You can expect intros that aren't just informative, but also relatable and engaging. They may even remind you of your beginnings in cybersecurity. I'll make complex topics feel accessible, highlight the human side of cybersecurity, and share insights that help the community grow stronger. 

At the end of the day, our work isn't just about threats, but about the humans working tirelessly to defend against them.

**The one big thing **

Talos has identified threats disguised as legitimate AI solution installers, including ransomware like CyberLock and Lucky\_Gh0$t, and a destructive malware called Numero. These threats highlight how malicious actors are leveraging the rise of AI to distribute harmful software. 

**Why do I care? **

Cybercriminals are targeting the trust and excitement around AI tools to deliver malware, which could affect anyone looking to adopt AI for personal or business use, putting their systems and data at risk. Understanding these threats helps you stay vigilant and avoid falling victim to such deceptive tactics. 

**So now what? **

Snort SIDs and ClamAV detections are available at the bottom of the blog post. Otherwise, always verify the source of any AI tools or software before downloading, use trusted cybersecurity solutions to protect your systems, and stay informed about emerging threats by keeping up with updates from reliable sources like Cisco Talos.

**Top security headlines of the week **

**MathWorks, Creator of MATLAB, Confirms Ransomware Attack**   
The attack dirsupted MathWorks' systems and online applications, but it remains unclear which ransomware group targeted the software company and whether they stole any data. (DarkReading) 

**Deepfakes, Scams, and the Age of Paranoia**   
This hit home, both as a jobseeker within the past year and a young(er) person who’s worried about her parents’ security. I may be able to parse AI portraits with six fingers and hair phasing through their clothes, but have you ever seen a convincing deepfake? Phew. (Wired) 

**Companies Warned of Commvault Vulnerability Exploitation**   
CISA says that the ongoing exploitation of a Commvault vulnerability that was targeted as a zero-day is likely part of a broader campaign against software-as-a-service (SaaS) solutions. (SecurityWeek) 

**US student agrees to plead guilty to hack affecting tens of millions of students**  
A Massachusetts student has agreed to plead guilty to federal charges relating to hacking and extorting one of the largest U.S. education tech companies. PI included names, addresses, phone numbers, Social Security numbers, medical information, and school grades. (TechCrunch)

**Can’t get enough Talos? **

**UAT-6382 exploits Cityworks zero-day vulnerability to deliver malware**  
Talos has found intrusions in enterprise networks of local governing bodies in the United States (U.S.), beginning January 2025 when initial exploitation first took place. **Read the blog here.**

**The day I found an APT group in the most unlikely place**   
In this Dark Reading Confidential episode, Talosian Vitor Ventura shares stories about the tricks he used to track down APTs, and the surprises discovered along the way. **Listen to the podcast here.**

**Upcoming events where you can find Talos **

*   Cisco Live U.S. (June 8 – 12) San Diego, CA 
*   NIRMA (July 28 – 30) St. Augustine, FL 
*   Black Hat USA (Aug. 2 - 7) Las Vegas, NV

**Most prevalent malware files from Talos telemetry over the past week **

**SHA 256: 9f1f11a708d393e0a4109ae189bc64f1f3e312653dcf317a2bd406f18ffcc507**   
MD5: 2915b3f8b703eb744fc54c81f4a9c67f   
VirusTotal: https://www.virustotal.com/gui/file/9f1f11a708d393e0a4109ae189bc64f1f3e312653dcf317a2bd406f18ffcc507    
Typical Filename: VID001.exe   
Claimed Product: N/A   
Detection Name: Win.Worm.Coinminer::1201 

**SHA 256: 59f1e69b68de4839c65b6e6d39ac7a272e2611ec1ed1bf73a4f455e2ca20eeaa**   
MD5: df11b3105df8d7c70e7b501e210e3cc3   
VirusTotal: https://www.virustotal.com/gui/file/59f1e69b68de4839c65b6e6d39ac7a272e2611ec1ed1bf73a4f455e2ca20eeaa   
Typical Filename: DOC001.exe   
Claimed Product: N/A   
Detection Name: Win.Worm.Coinminer::1201 

**SHA256:3294df8e416f72225ab1ccf0ed0390134604bc747d60c36fbb8270f96732e341**   
MD5: b6bc3353a164b35f5b815fc1c429eaab   
VirusTotal: https://www.virustotal.com/gui/file/3294df8e416f72225ab1ccf0ed0390134604bc747d60c36fbb8270f96732e341   
Typical Filename: b6bc3353a164b35f5b815fc1c429eaab.msi   
Claimed Product: n/a    
Detection Name: Simple\_Custom\_Detection 

**SHA 256: a31f222fc283227f5e7988d1ad9c0aecd66d58bb7b4d8518ae23e110308dbf91**     
MD5: 7bdbd180c081fa63ca94f9c22c457376    
VirusTotal: https://www.virustotal.com/gui/file/a31f222fc283227f5e7988d1ad9c0aecd66d58bb7b4d8518ae23e110308dbf91   
Typical Filename: c0dwjdi6a.dll    
Claimed Product: N/A     
Detection Name: Trojan.GenericKD.33515991

A new author has appeared

Cisco Talos has uncovered new threats, including ransomware like CyberLock and Lucky_Gh0$t, and a destructive malware called Numero, all disguised as legitimate AI tool installers to target victims.

Thursday, May 29, 2025 06:00

*   Cisco Talos has discovered new threats, including the ransomware CyberLock, Lucky\_Gh0$t, and a newly-discovered malware we call “Numero,” all of which masquerade as legitimate AI tool installers. 
*   CyberLock ransomware, developed using PowerShell, primarily focuses on encrypting specific files on the victim's system. The threat actor deceitfully claims in the ransom note that the payments will be allocated for humanitarian aid in various regions, including Palestine, Ukraine, Africa and Asia. 
*   Lucky\_Gh0$t ransomware is yet another variant of the Yashma ransomware, which is the sixth iteration of the Chaos ransomware series, featuring only minor modifications to the ransomware binary. 
*   The newly-identified destructive malware, Numero, affects victims by manipulating the graphical user interface (GUI) components of their Windows OSs, rendering systems completely unusable. 

AI has increasingly proliferated across various business verticals, leading to a transformation of industries through automation, data-driven decision-making and enhanced customer engagements. However, as AI continues to propel multiple industry sectors forward, malicious actors are exploiting its popularity by distributing a range of malware disguised as AI solutions’ installers and tools.  

Threat actors are employing a variety of techniques and channels to distribute these fraudulent installers, including SEO-poisoning tactics to manipulate search engine rankings and cause their malicious websites or download links to appear at the top of search engine results, as well as platforms such as Telegram or social media messengers. 

As a result, unsuspecting businesses in search of AI solutions may be deceived into downloading counterfeit tools in which malware is embedded. This practice poses a significant risk, as it not only compromises sensitive business data and financial assets but also undermines trust in legitimate AI market solutions. Therefore, organizations and users must exercise extreme caution, meticulously verify sources, and rely exclusively on reputable vendors to avoid falling prey to these threats. 

Talos has recently uncovered multiple threats masquerading as AI solutions being circulated in the wild, including the CyberLock and Lucky\_Gh0$t ransomware families, along with a newly discovered destructive malware, dubbed “Numero.” The legitimate versions of these AI tools are particularly popular within the B2B sales domain and the technology and marketing sectors, indicating that individuals and organizations in these industries are particularly at risk of being targeted by these malicious threats. 

**CyberLock ransomware **

Talos observed a threat actor creating a lookalike fake AI solution website with the domain ‘novaleadsai\[.\]com’, likely masquerading as the original website domain ‘novaleads.app’, a lead monetization platform designed to help businesses maximize the value of their leads through various services and performance-based models. 

Figure 1. Fake website advertising the AI tool. 

On the fake website, the actor persuades users to download the product with an offer of free access to the tool for the first 12 months, followed with a monthly subscription of $95. The threat actor also used an SEO manipulation technique that made their fake website appear in the top search results for online search engines.   

When a user downloads the fake AI product as a ZIP archive, it contains a .NET executable with the file name ‘NovaLeadsAI.exe’. The executable was compiled on Feb. 2, 2025, which is on the same day the fake domain ‘novaleadsai\[.\]com’ was created.  

The ‘NovaLeadsAI.exe’ file is the loader that has the CyberLock ransomware PowerShell script embedded as the resource file. When the victim runs the loader executable, it deploys the ransomware. 

Figure 2. Snippet of the CyberLock ransomware loader. 

**CyberLock ransom note **

The CyberLock ransomware appeared to be operating as early as Feb. 2025. The ransom note claims that the threat actor has obtained full access to sensitive business documents, personal files and confidential databases, demanding a hefty ransom in exchange for decryption keys. Victims are instructed to communicate with the threat actor by emailing ‘cyberspectreislocked@onionmail\[.\]org’. 

The CyberLock threat actor demands that the USD $50,000 ransom be paid exclusively in Monero (XMR) cryptocurrency and employs psychological tactics by falsely claiming that the ransom payments will be used for humanitarian aid in regions like Palestine, Ukraine, Africa and Asia. The actor splits the payment into two separate wallets, complicating defenders' tracking efforts. 

The ransom note is structured to intimidate and manipulate victims by threatening to expose stolen data if payment is not made within three days. However, Talos did not see any evidence of data exfiltration functionality within the ransomware code. 

Figure 3. CyberLock ransom note.

**CyberLock, the PowerShell ransomware **

CyberLock ransomware is written in PowerShell, embedded with the CSharp code and delivered to the victims as an embedded resource of the .NET loader. 

When CyberLock is executed, it initially uses the functions GetConsoleWindow from kernel32.dll and ShowWindow from user32.dll to hide the PowerShell window. Then it generates a secret by decrypting the encrypted public key and uses it to derive the AES key and IV during the encryption process. 

Figure 4. Snippet of CyberLock ransomware. 

CyberLock has the capability to elevate privileges and re-execute itself with administrative privileges if it is not already running in an elevated context.   

Figure 5. Snippet of CyberLock ransomware. 

CyberLocker enumerates folders and files of the logical partitions with the labels ‘C:\\’, ‘D:\\’ and ‘E:\\’. It encrypts the targeted files using AES and appends the file extension ‘.cyberlock’ to the encrypted files.  

Figure 6. Snippet of CyberLock ransomware. 

The targeted file extensions and the categories are shown below: 

Category

File Extensions

Text Documents

.txt, .doc, .docx, .odt, .rtf, .md, .rst, .tex, .sty

Spreadsheets

.xls, .xlsx, .ods, .csv, .tsv

Presentations

.ppt, .pptx, .odp, .potx, .ppsx

PDF & eBooks

.pdf, .pdfx, .epub, .mobi, .azw, .azw3, .chm, .hlp

Images

.jpg, .jpeg, .png, .gif, .bmp, .tiff, .raw, .svg, .jfif, .ico, .webp

Audio

.mp3, .wav, .ogg, .aac, .flac, .m4a, .m4b, .caf, .mp3g

Video

.avi, .mp4, .mov, .mkv, .wmv, .webm, .3gp, .flv, .m4v, .vob, .mts, .m2ts, .ts, .mxf, .divx, .mpeg, .mpg, .ram, .rm

Archives & Disk Images

.zip, .rar, .7z, .tar, .gz, .xz, .tar.gz, .tar.bz2, .iso, .iso9660, .img, .dmg, .cdr, .zipx, .cab, .zpaq, .seam, .rar5

Executables & Scripts

.exe, .bat, .cmd, .sh, .ps1, .vbs, .js, .appx, .apk, .ipa, .deb, .rpm, .whl

Code & Programming

.html, .css, .scss, .xml, .json, .yaml, .cfg, .sql, .pl, .rb, .py, .lua, .h, .c, .cpp, .m, .swift, .java, .asm, .psm1

Database Files

.sql, .mdb, .accdb, .db, .sqlite, .sqlitedb, .db3, .sqlite3

System & Config

.log, .bak, .tmp, .swp, .ini, .plist, .xmlrpc, .dsk, .xcv

Fonts

.ttf, .otf, .woff, .woff2, .eot, .pfb

Design & Graphics

.ai, .psd, .indd, .eps, .fla, .swf

Backup & Virtual Machine

.vhd, .vmdk, .qcow2, .gho, .vpb

GIS & Maps

.gpx, .kml, .shp

Other Files

.torrent, .bup, .ifo, .bin, .dll, .msi, .sys, .qif, .pages, .key, .numbers, .rdata, .seed, .3dxml, .kdbx

After encrypting the targeted files, CyberLock creates a ransom note on the victim machine desktop with the file name ‘ReadMeNow.txt’. Ransom note contents are written into it from the embedded strings in the ransomware PowerShell script. 

Talos observed that the ransomware actor sets a wallpaper to the victim machine’s desktop after dropping the ransom note. The threat actor downloads a header image from a cybersecurity organization’s blog post to the victim machine user profile applications temporary folder. They configure the path of the downloaded image to the registry key “Wallpaper” and enable the wallpaper through PowerShell commands. Talos not fully certain of the actor’s motive for setting the victim machine’s desktop wallpaper to a security research blog post header image. 

Figure 7. Snippet of CyberLock ransomware. 

Figure 8. Sample blog post header wallpaper. 

Finally, CyberLock uses the living-off-the-land binary (LoLBin) ‘cipher.exe’ with the ‘/w’ option to erase free space on the victim's hard drive partitions, hindering forensic recovery of deleted files. 

Figure 9. Command execution to wipe the hard drive free space. 

‘Cipher.exe’ is a built-in Windows command-line tool for managing file and folder encryption. One of its features allows users to prevent recovery of deleted files by overwriting free space with the ‘/w’ option. This was designed by Microsoft for legitimate purposes, such as securely wiping disks before reallocating them or complying with data protection laws to ensure sensitive data is unrecoverable by unauthorized parties.  

Threat actors often misuse this feature to eliminate their malicious footprints or permanently delete files from victim machines. This technique was previously utilized by a Russian APT in their attacks, as noted by Volexity researchers. However, Talos has not observed any indication that this activity is related to the activity described in prior reporting. 

**Lucky\_Gh0$t ransomware as fake ChatGPT installer  **

Talos discovered a threat actor distributing Lucky\_Gh0$t ransomware in the wild, archived in a self-extracting archive (SFX) ZIP installer with the file name ‘ChatGPT 4.0 full version - Premium.exe’. 

The malicious SFX installer included a folder that contained the Lucky\_Gh0$t ransomware executable with the filename ‘dwn.exe’, which imitates the legitimate Microsoft executable ‘dwm.exe’. The folder also contained legitimate Microsoft open-source AI tools that are available on their GitHub repository for developers and data scientists working with AI, particularly within the Azure ecosystem. The threat actor’s intention in including the legitimate tools in the SFX archive is likely to evade the anti-malware file scanners detections by masquerading as a legitimate package.  

The SFX script executes the ransomware when a victim runs the malicious SFX installer file. 

Figure 10. Malicious SFX executable contents. 

Lucky\_Gh0$t ransomware is the Yashma ransomware variant with most features unchanged, including the evasion techniques, deleting the volume shadow copies and backups, and AES-256 and RSA-2048 encryption techniques. Talos observed a few minor modifications in the Lucky\_Gh0$t binary with targeted file size limits that are to be considered by the ransomware during encryption. 

Lucky\_Gh0$t targets files on the victim machine that are approximately less than 1.2GB in size and encrypts the files with the RSA-encrypted AES key, appending a 4-digit random alphanumeric characters as the file extension. The targeted files category for encryption include: 

*   Text, code and config files 
*   Microsoft Office and Adobe files 
*   Media formats and images 
*   Archives and installers 
*   Backup and database files 
*   Android package kit, Java Server Pages and Active server pages 
*   Certificate files 
*   Visual Studio Solutions and PostScripts 

Figure 11. Lucky\_Gh0$t encryption function for files less than 1.2GB. 

For the targeted files with a size larger than 1.2GB, the ransomware creates a new file the same size of the original file and writes a single character “?” as the file content. It appends a 4-digit random alphanumeric character file extension to the new file and deletes the original file, exhibiting destructive behavior. 

Figure 12. Lucky\_Gh0$t encryption function for files larger than 1.2GB. 

Lucky\_Gh0$t ransomware provides a personal ID to the victims in their ransom note. For further communication regarding ransom payment and decryption, it instructs the victims to contact the threat actor using a secure messenger platform at ‘getsession\[.\]org’ with a unique session ID.

Figure 13. Lucky\_Gh0$t ransom note.

Talos recently discovered a new destructive malware in the wild that we call “Numero,” designed to imitate the AI video creation tool installer, InVideo AI. InVideo AI is an online platform widely used for marketing videos, social media content, explainer videos and presentations. The threat actor impersonates the product and the organization names in the malicious file’s metadata. 

Figure 14. A fake installer execution flow running the payload Numero. 

The fake installer is a dropper containing a malicious Windows batch file, VB script and the Numero executable with the file name ‘wintitle.exe’. When the victim runs the fake installer, it drops the malicious components in a folder at the local user profile’s application temporary folder. Then it executes the dropped Windows batch file through Windows shell in an infinite loop. It first runs the Numero malware and then halts the execution for 60 seconds by executing the VB script through cscript.  

After resuming the execution, the batch file terminates the Numero malware process and restarts its execution. By implementing the infinite loop in the batch file, the Numero malware is continuously run on the victim machine. 

Figure 15. Malicious Windows bat loader. 

Numero's behavior is consistent with window manipulator malware. Numero is a 32-bit windows executable written in C++ and was compiled on Jan. 24, 2025.  

Numero evades analysis by checking the process handles of various malware analysis tools and debuggers including IDA, x64 debugger, x32debugger, ollydbg, scylla, windbg, reshacker, ImportREC, Immunity debugger, Zeta debugger and Rock debugger. 

Figure 16. Snippet of the Numero function and the malicious thread. 

Numero malware creates and executes the thread in an infinite loop. The thread code interacts with the Windows GUI and manipulates the victim’s desktop window using the Windows APIs GetDesktopWindow, EnumChildWindows and SendMessageW. It monitors the victim machine desktop window continuously and hooks to the child window created in the victim desktop. Numero overwrites the window title, buttons and contents with the numeric string ‘1234567890’, corrupting the victim machine to become unusable.  

Figure 17. Corrupted Windows Run terminal. 

**Coverage  **

Ways our customers can detect and block this threat are listed below.  

Cisco Secure Endpoint (formerly AMP for Endpoints) is ideally suited to prevent the execution of the malware detailed in this post. Try Secure Endpoint for free here.  

Cisco Secure Email (formerly Cisco Email Security) can block malicious emails sent by threat actors as part of their campaign. You can try Secure Email for free here.  

Cisco Secure Firewall (formerly Next-Generation Firewall and Firepower NGFW) appliances such as Threat Defense Virtual, Adaptive Security Appliance and Meraki MX can detect malicious activity associated with this threat.  

Cisco Secure Network/Cloud Analytics (Stealthwatch/Stealthwatch Cloud) analyzes network traffic automatically and alerts users of potentially unwanted activity on every connected device.  

Cisco Secure Malware Analytics (Threat Grid) identifies malicious binaries and builds protection into all Cisco Secure products.  

Cisco Secure Access is a modern cloud-delivered Security Service Edge (SSE) built on Zero Trust principles.  Secure Access provides seamless transparent and secure access to the internet, cloud services or private application no matter where your users work.  Please contact your Cisco account representative or authorized partner if you are interested in a free trial of Cisco Secure Access.  

Umbrella, Cisco’s secure internet gateway (SIG), blocks users from connecting to malicious domains, IPs and URLs, whether users are on or off the corporate network.   

Cisco Secure Web Appliance (formerly Web Security Appliance) automatically blocks potentially dangerous sites and tests suspicious sites before users access them.   

Additional protections with context to your specific environment and threat data are available from the Firewall Management Center.  

Cisco Duo provides multi-factor authentication for users to ensure only those authorized are accessing your network.   

Open-source Snort Subscriber Rule Set customers can stay up to date by downloading the latest rule pack available for purchase on Snort.org.  

Snort SIDs for the threats are: 

*   Snort 2: 64901, 64902, 64899, 64900, 64897, 64898, 64896 
*   Snort 3: 301207, 301206, 301205 

ClamAV detections are also available for this threat:   

*   Ps1.Ransomware.CyberLock-10045054-0 
*   Win.Dropper.CyberLock-10045058-0 
*   Win.Dropper.LuckyGhost-10045078-0 
*   Win.Ransomware.LuckyGhost-10045080-0 
*   Win.Loader.Numero-10045084-0 
*   Win.Dropper.Numero-10045088-0 
*   Win.Malware.Numero-10045090-0 
*   Win.Malware.Numero-10045093-0 

**Indicators of Compromise **

IOCs for this threat can be found in our GitHub repository here.

Cybercriminals camouflaging threats as AI tool installers

Talos analyzed six months of PowerShell network telemetry and found that rare domains are over three times more likely to be malicious compared to frequently contacted ones.

Friday, May 23, 2025 06:00

_By Darin Smith and John Arneson_

*   Cisco Talos reviewed six months of network connection telemetry logs spanning June 1, 2024 – Dec. 31, 2024, containing 3,220,829 log events and 742 unique base domains, to explore if domains that PowerShell rarely contacts are more likely to be malicious. 
*   Key findings reveal that the odds of a rare domain being malicious were 3.18 times higher than for frequently contacted domains (95% CI: 0.39–25.9), suggesting a trend towards higher risk in rare domains. 
*   Notably, the non-rare domain ‘githubusercontent.com’ was flagged as malicious due to activity from its subdomain ‘raw.githubusercontent.com’. This is an example of why subdomains should be considered when looking for malicious network traffic, especially for cloud services where the service itself is legitimate, but the content hosted on it is not guaranteed to be.  

**Research Methodology ****Hypothesis**

At a sufficiently high volume of telemetry, domain names that PowerShell rarely connects to are more likely to be malicious than domains that are frequently connected to, regardless of PowerShell module. 

**Data Collection **

Talos queried telemetry for PowerShell network connection logs from a time period of June 1, 2024 to Dec. 31, 2024. This dataset included the following processes: ‘powershell.exe’, ‘powershell studio.exe’, ‘powershell\_ise.exe’, ‘powershelltools.exe’, ‘powershelltoolsx64.exe’, ‘pwsh’, and ‘pwsh.exe’. All of these processes are different versions of PowerShell. Talos excluded non-public top-level domains (TLDs), such as internal domains, to focus on external connections.  

**Data Processing**

Using the tldextract library, Talos extracted base domains (e.g., ‘automox.com’ from ‘api.automox.com’), resulting in 742 unique base domains. Rarity was defined as an average of ≤5 average contacts per full domain, calculated by dividing the total contacts by the number of unique full domains per base domain. This threshold identified 550 rare domains (74.1% of the total).  

**Threat Intelligence and Manual Review **

Talos assessed domain reputation using ReversingLabs (RL), which flagged a domain as malicious if any third-party source indicated so. To mitigate false positives (e.g., ‘adobe.com’), 29 domains were manually reviewed and overridden as benign, and their process arguments were documented. For subdomains such as ‘raw.githubusercontent.com’ under ‘githubusercontent.com’, the process arguments in those logs were manually reviewed, flagging 5 out of 10 connections as malicious based on commands like downloading PowerSploit or executing Invoke-Mimikatz, ensuring comprehensive threat detection.

**Findings & Analysis ****Domain Contact Distribution **

The distribution of contacts was heavily skewed: 

*   **Percentiles**: 60th percentile at 5.0 contacts, 90th at 82.0, 95th at 321.55, and 99th at 7,925.87 
*   **Top Domains**: ‘automox.com’ (2,282,308 contacts), ‘launchdarkly.com’ (493,812), and ‘amazonaws.com’ (166,536) accounted for most activity.
    *   Automox is a service for automated endpoint configuration and patch management.
    *   LaunchDarkly is a software development platform for managing feature flags and context-aware targeting of features.
    *   Amazon Web Services (AWS) is the largest cloud service provider. 
*   **Rare Domains**: 550 of 742 domains fell into the rare category.

Figure 1. Cumulative distribution of domain contact frequencies. 

**Malicious Domain Statistics **

*   **Rare Domains**: 9 malicious out of 550 (1.64%, 95% CI: 0.86%–3.08%)
*   **Non-Rare Domains**: 1 malicious out of 192 (0.52%, 95% CI: 0.09%–2.89%), notably ‘githubusercontent.com’
*   **Odds Ratio**: 3.18 (95% CI: 0.39–25.9), indicating a trend towards higher risk in rare domains, though not statistically significant (chi-square p=0.4291, Fisher’s exact p=0.4668), likely due to small sample sizes (9 rare, 1 non-rare) 

Figure 2. Malicious rates by domain rarity.

**Case Study: githubusercontent.com **

The non-rare domain ‘githubusercontent.com’ (38 contacts, 2 full domains: ‘raw.githubusercontent.com’ and ‘objects.githubusercontent.com’, average 19.00 contacts per full domain) was flagged as malicious due to 5 manually identified malicious contacts from ‘raw.githubusercontent.com’. These contacts involved potentially malicious PowerShell commands, such as downloading and executing scripts like PowerSploit or Invoke-Mimikatz. The other subdomain, ‘objects.githubusercontent.com’ (28 contacts), showed no malicious activity. This finding illustrates that even frequently contacted domains can host malicious subdomains, emphasizing the need for subdomain-level analysis in threat detection.

**Comparison to other Processes **

Another research question investigated was how the domains contacted by other similar processes would compare to those contacted by PowerShell. For the purposes of this research, Talos chose the following processes for analysis: 

*   ‘rundll32.exe’ 
*   Python (including macOS and Windows versions) 
*   ‘cmd.exe’ 
*   ‘cscript.exe’ 
*   ‘wscript.exe’ 
*   ‘bash’ 
*   ‘zsh’

These processes are primarily other command line or script interpreters, as well as ‘rundll32.exe’, which allows executing Dynamically Linked Libraries (DLLs) from the command line.

When the same heuristics as were utilized for PowerShell were applied to the domains contacted by these processes, the results varied somewhat. Across 156,203 total connection records for ‘rundll32.exe’, 940 unique domains were contacted. Of these, 722 of these domains were “rare,” using the same heuristic applied to PowerShell (i.e., they were contacted at most five times). Only one of the domains contacted was found to be malicious, either among the rare domains or the non-rare domains.

Similarly, among 795,346 total connection records for Python, 825 unique domains were contacted and 616 were rare using the same criteria. None of the rare domains were malicious, while 1 of the non-rare domains was. The processes cscript, cmd, zsh and csh had similar results, with no or single digit numbers of malicious domains contacted. However, wscript was much more interesting. It had a much smaller amount of total utilization in the dataset investigated, with just 6,936 connection events and 82 unique domains contacted. Of these, 58 domains were rare (or roughly 71%), and 5 were found to be malicious.

**Recommendations **

*   **Prioritize Rare Domains**: Security teams should focus investigations on rare domains due to their higher likelihood of being malicious, despite statistical non-significance. This finding applies primarily to PowerShell and wscript among the processes considered in this research.  
*   **Subdomain Analysis**: For frequently contacted domains, analyze subdomains and process arguments to detect malicious activity, as demonstrated with ‘githubusercontent.com’. 
*   **Integrate Manual Review**: Combine automated threat intelligence with manual reviews to reduce false positives and identify nuanced threats, particularly in high-contact domains. 
*   **Investigate Anomalous Utilization of ‘wscript.exe’:** Some environments may still commonly utilize wscript. However, this research suggests that in environments where it is rare, it has the highest likelihood to be used to connect to malicious domains of the processes researched.

**Future Work **

This research presents several opportunities for future research. One opportunity is temporal analysis to determine if there were time-based patterns for contacting domains, and if so, determining if these patterns could be used to identify malicious activity. This could potentially include seeing increased contacts of malicious domains during weekends or off-hours. Time-series analysis could be applied to the data to test this hypothesis.

Another opportunity is the behavioral analysis of process arguments, focusing on identifying recurring patterns tied to malicious activity, such as downloading PowerShell scripts from a remote host, or exfiltration of data. This could be used to refine the current rarity to malicious correlation of 1.64% for rare domains versus 0.52% for non-rare domains. This may spotlight behavioral red flags and give actionable insights for more precision detection logic.

Finally, future research can develop a risk scoring system that integrates multiple factors such as contact frequency, malicious rate, TLDs and even ReversingLabs’ network threat intelligence. This can provide a scalable and practical tool for security teams to prioritize high-risk domains, whether rare or non-rare like ‘githubusercontent.com’. This builds on the current analysis but also paves the way for more robust, data-driven strategies to combat threats, ensuring this research delivers lasting value to the security community.

Scarcity signals: Are rare activities red flags?

Hazel observes that cybercriminals often fumble teamwork, with fragile alliances crumbling over missed messages. Plus, how UAT-6382 is exploiting Cityworks and what you can do to stay secure.

Thursday, May 22, 2025 14:00

Welcome to this week’s edition of the Threat Source newsletter. 

Talos recently published research into how threat actors are increasingly teaming up across the attack chain. Each group handles a slice of the operation, passing the breach along like a relay baton. 

It’s a concerning trend — one that we believe calls for rethinking traditional threat modeling. But one thing stood out to me while reading: cybercriminals are often terrible at teamwork. 

What if the ransomware affiliate is waiting on credentials that never arrive? The access broker sells a foothold, but the tooling meant to exploit it isn’t ready, doesn’t work in the target environment or never shows up at all? 

Ghosting isn’t limited to dating apps or job interviews (and if you’ve been through six interview rounds and still heard nothing, I see you). Cybercriminals flake too — whether it’s bad timing, better targets, internal drama… or maybe they just went to get a haircut (an actual complaint that a Conti member made about a fellow actor not showing up). 

In this compartmentalized model, the threat chain becomes a fragile supply line, stitched together in real time. Efficient, yes — but brittle. If one actor drops out, the whole operation can unravel. And let’s not pretend there’s honour among cybercriminals. They're opportunists. What’s to stop a broker from selling the same credentials to multiple buyers? Or backing out entirely if a better offer lands? 

Of course, this ecosystem isn’t monolithic. Some groups run like structured businesses — access brokers, malware builders, “customer” (aka victim) services, the works. Others are looser, relying on whoever turns up in their DMs with access for sale. It’s the latter where ghosting seems more likely. In organised crews, a flaky broker risks reputational damage. In the freelance underworld, it’s just Tuesday.  

Oof, I didn’t mean to knock freelancers there. Just, you know, _those_ ones… 

History suggests fallouts are inevitable. Conti's collapse, as Wired reported, started with a single angry post and spiraled into a full on leak about poor performance records: 

> _“I have 100 people here, half of them, even 10 percent, do not do what they need.”_  

> _\- Stern (or Demon), former Conti CEO_ 

LAPSUS$ imploded under its own infighting. One REvil affiliate even ranted on a cybercrime forum like a scammed eBay buyer. 

To twist a familiar phrase: compartmentalized threats are only as strong as their weakest link. What if that link has poor communication skills, no follow-through and a serious case of commitment issues?

**The one big thing **

In Talos’ most recent blog post, we shared that UAT-6382, Chinese-speaking threat actors, have exploited Cityworks, a widely-used asset management system, through a remote code execution vulnerability (CVE-2025-0994). The actors are deploying advanced malware for long-term persistence and control. 

**Why do I care? **

UAT-6382 is not only exploiting this vulnerability, but they're also employing sophisticated tools like web shells, Rust-based malware loaders, and frameworks like Cobalt Strike to burrow deep into systems. This could lead to data breaches and operational downtime. 

**So now what? **

While the intrusions we mentioned in the blog have been contained, exploitation may be continuing in the wild. Use the indicators of compromise (IOCs) listed in the blog to scan your environment.

**Top security headlines of the week **

**NATO-Flagged Vulnerability Tops Latest VMware Security Patch Batch**   
VMware patches flaws that expose users to data leakage, command execution and denial-of-service attacks. No temporary workarounds available.  (SecurityWeek) 

**NIST's 'LEV' Equation to Determine Likelihood a Bug Was Exploited**   
The new equation, introduced by the National Institute of Standards and Technology (NIST), aims to offer a mathematical likelihood index that could be a game-changer for SecOps teams and vulnerability patch prioritization. (Dark Reading) 

**Kettering Health hit by system-wide outage after ransomware attack**   
Kettering Health, a healthcare network that operates 14 medical centers in Ohio, was forced to cancel inpatient and outpatient procedures following a cyberattack that caused a system-wide technology outage. (BleepingComputer)

**Can’t get enough Talos? **

*   Duping Cloud Functions: An emerging serverless attack vector 
*   Talos Takes: Inside the Kill Chain: Compartmentalized Threat Modeling Explained 

**Upcoming events where you can find Talos **

*   BotConf (May 20 – 23) Angers, France  
*   Cisco Live U.S. (June 8 – 12) San Diego, CA 
*   NIRMA (July 28 – 30) St. Augustine, FL 
*   Black Hat USA (August 2 - 7) Las Vegas, NV

**Most prevalent malware files from Talos telemetry over the past week **

**SHA 256: 9f1f11a708d393e0a4109ae189bc64f1f3e312653dcf317a2bd406f18ffcc507**   
MD5: 2915b3f8b703eb744fc54c81f4a9c67f   
VirusTotal: https://www.virustotal.com/gui/file/9f1f11a708d393e0a4109ae189bc64f1f3e312653dcf317a2bd406f18ffcc507   
Typical Filename: VID001.exe   
Claimed Product: N/A   
Detection Name: Win.Worm.Coinminer::1201 

**SHA 256: a31f222fc283227f5e7988d1ad9c0aecd66d58bb7b4d8518ae23e110308dbf91**     
MD5: 7bdbd180c081fa63ca94f9c22c457376     
VirusTotal: https://www.virustotal.com/gui/file/a31f222fc283227f5e7988d1ad9c0aecd66d58bb7b4d8518ae23e110308dbf91/details    
Typical Filename: IMG001.exe    
Detection Name: Simple\_Custom\_Detection 

**SHA256: 47ecaab5cd6b26fe18d9759a9392bce81ba379817c53a3a468fe9060a076f8ca**    
MD5: 71fea034b422e4a17ebb06022532fdde    
VirusTotal: https://www.virustotal.com/gui/file/47ecaab5cd6b26fe18d9759a9392bce81ba379817c53a3a468fe9060a076f8ca/details    
Typical Filename: VID001.exe    
Claimed Product: N/A    
Detection Name: Coinminer:MBT.26mw.in14.Talos 

**SHA 256: c67b03c0a91eaefffd2f2c79b5c26a2648b8d3c19a22cadf35453455ff08ead0**   
MD5: 8c69830a50fb85d8a794fa46643493b2    
Typical Filename: AAct.exe    
Claimed Product: N/A    
Detection Name: PUA.Win.Dropper.Generic::1201

Ghosted by a cybercriminal

Talos has observed exploitation of CVE-2025-0994 in the wild by UAT-6382, a Chinese-speaking threat actor, who then deployed malware payloads via TetraLoader.

UAT-6382 exploits Cityworks zero-day vulnerability to deliver malware

Cisco Talos built on Tenable’s discovery of a Google Cloud Platform vulnerability to uncover how attackers could exploit similar techniques across AWS and Azure.

Tuesday, May 20, 2025 06:00

**Summary and background**

Google Cloud Platform (GCP) Cloud Functions are event-triggered, serverless functions that automatically scale and execute code in response to specific events like Hypertext Transfer Protocol (HTTP) requests or data changes. Tenable Research published an article discussing a vulnerability they discovered within GCP’s Cloud Functions serverless compute service and its Cloud Build continuous integration and continuous delivery or deployment (CI/CD) pipeline service.

“When a GCP user creates or updates a Cloud Function, a multi-step backend process is triggered,” Tenable author Liv Matan writes. “This process, among other things, attaches a default Cloud Build service account to the Cloud Build instance that is created as part of the function’s deployment.” This default Cloud Build Service Account (SA) previously gave users excessive Cloud Function permissions. An attacker who has gained the ability to create or update a cloud function could utilize the function’s deployment process to escalate privileges to the default Cloud Build service account or assign a higher privileged SA. Google has since partially addressed Tenable’s discovery to ensure the default Cloud Build service account no longer provides users with excessive permissions.

Based on Tenable’s research, Cisco Talos conducted a series of offensive tests within Cisco’s Google Cloud Platform (GCP) to identify additional threats that may affect customer environments.

During its research, Talos discovered that the technique Tenable identified could be adapted to perform other malicious activities. By implementing different malicious console commands into the Node Package Manager (NPM) ‘package.json’ file used in this technique, threat actors could execute behaviors such as environment enumeration.

Talos furthered this research by attempting to replicate similar behaviors in Amazon Web Services (AWS) and Microsoft Azure to determine if these techniques could be employed to perform similar malicious activities in other cloud-based environments.

**Research****Prerequisites**

To utilize this attack vector, certain prerequisites must be met. Talos set up a Debian Linux server within the GCP environment with Node Package Manager (NPM) and Ngrok installed. However, the virtual machine for this research can be created in any cloud environment.

After installing NPM and Ngrok, Talos configured both tools to function as intended.

Once NPM and Ngrok were configured, a Python server was created to output the data received from the cloud function.

With NPM, Ngrok, and the Python server set up and configured, the next step was to create and modify the NPM package.

Talos then replaced the content of the package.json file with the following code:

Finally, once all the necessary files are created and configured, Talos set up the environment to visually display the data output from deploying the functions. To achieve this, Talos activated both the Ngrok server and the Python server created earlier.

To replicate the GCP behavior discussed in Tenable’s article, Talos created/updated an SA with function build and cloud build permissions. This SA was then assigned to the GCP Cloud Run Function to allow the code to be executed with privileged access.

Once the servers and service accounts were online and configured to receive and output data, the emulation of the behavior could begin.

**Emulation**

With the package.json file configured to be utilized by the build function, Talos began emulating the technique described in Tenable’s research article.

The first step in Talos’ replication involved the utilizing a misconfigured GCP function to extract the default Cloud Build service account token. To initiate this process, the "malicious" package.json was updated on the virtual machine, ensuring that it contains code similar to that used by Tenable.

Once the package.json file was modified as desired, it needed to be published to the public NPM registry. To do this, Talos executed the following command:

With the package.json file uploaded to the NPM public registry, it was time to deploy the GCP Cloud Run Function so that the package.json can execute the provided code. To do this, the user must to navigate to their GCP Cloud Run Functions page and select or create a Cloud Run Function, ensuring it is assigned a service account with Cloud Build permissions.

Figure 1. Google Cloud Run Function displaying the assigned service account.

As Talos created or selected our existing GCP Cloud Run Function, we navigated to the source page of the cloud function. Here, Talos modified the package.json file to install the malicious package uploaded to NPM.

Figure 2. Google Cloud Run Function’s Source page.

Once Talos updated the package.json file with the correct name and version of the NPM package, we selected "Deploy" or "Save and Redeploy" to initiate the build process. During this process, the function sends the requested data to the Ngrok server, which was then output on the Python server.

Talos confirmed that the exfiltration of GCP service account access tokens can no longer be achieved using this method, due to Google's response and patching of the issue. We further verified this by executing the same command provided to our NPM-uploaded package.json from a separate virtual machine. The command executed successfully, confirming our suspicion that this specific technique for obtaining privileged service account tokens has been patched out.

**Original Research**

Cisco Talos' research extended Tenable’s original behavior concept by applying it to other cloud environments through modifications to their respective cloud services. AWS Lambda and Azure Functions are serverless compute services that allow users to run code without provisioning or managing servers. By creating a Lambda function or an Azure function with a Node.js 20.x runtime, a package.json file can be created with dependencies set to execute a malicious package uploaded to NPM’s public repository. These malicious packages may contain harmful console commands that provide a threat actor with valuable enumeration information.

Although this specific vector of threat actor behavior is no longer possible, other commands have proven useful in providing adversaries with valuable enumeration capabilities. These commands can be used on cloud platforms beyond GCP Cloud Build Function, such as AWS Lambda and Azure Functions.

Some examples of the types of enumeration a threat actor can perform using this method include the following.

**ICMP Discovery**

Internet Control Message Protocol (ICMP) Discovery is utilized to gather information about network devices and their configurations. By analyzing ICMP responses, adversaries can infer the network's structure, including the presence of routers, gateways, and the pathways between devices. This information can be crucial for planning attacks.

**Existence of .dockerenv**

Identifying the presence of a .dockerenv file indicates that a process is running inside a Docker container. By checking for this file, threat actors can confirm whether they are operating within a Docker environment. This information can influence their selection of tools and techniques, as containers often possess different security boundaries compared to host systems.

**CPU Scheduling**

Enumerating CPU Scheduling provides detailed scheduling and status information about the process with process identifier (PID) 1, which is typically the init system or main process in a containerized environment. Threat actors can determine the init system in use, such as systemd or sysvinit. This information helps them understand the system's configuration and identify potential vulnerabilities associated with the specific init system.

CPU Scheduling Data Output Plain Text

**Control Group Container ID**

Enumerating Control Group Container ID provides detailed information about current mount points. Threat actors can use this information to identify critical or sensitive filesystems that might be targeted for data exfiltration. By examining mount options, they can look for insecure configurations, such as filesystems mounted with exec permissions in directories where malicious binaries could be introduced. In containerized environments, understanding mount namespaces can aid in developing container escape techniques, enabling attackers to break out of the container and access the host system.

Control Group Container ID Plain Text 1 & Control Group Container ID Plain Text 2

**Initial Server Overview**

For Initial Server Overview enumeration, combining the following commands provides comprehensive details about the system's kernel, architecture and distribution, which are critical for understanding the environment and planning potential exploits. Knowing the exact OS and kernel version enables threat actors to choose the most effective exploits, as many vulnerabilities are version-specific.

**User and Permission Enumeration**

The following User and Permission commands provides insights into user accounts, privileges and group memberships, which are crucial for planning privilege escalation and lateral movement within a system.

**Network Discovery**

The following Network and Discovery commands help gather detailed insights into the system's operating environment and network setup, which can be used to identify vulnerabilities and plan attacks.

**Detailed System Commands**

The ‘cat /etc/os-release’ command reveals the operating system distribution and version. Knowing the exact OS helps attackers identify specific vulnerabilities and tailor their exploits to the target's environment.

**User Related Commands**

The ‘/etc/shadow’ file contains hashed passwords for user accounts, which, if accessed, can be used to crack passwords and gain elevated access to the system.

User Related Commands Data Output Plain Text

**AWS Lambda Functions**

The following example demonstrates Talos using the same commands previously mentioned within a Google Cloud Platform (GCP) environment, now applied in an Amazon Web Services (AWS) environment using Lambda functions. This illustrates that the method utilized by the Tenable lab can be adapted for other cloud-based environments, such as AWS.

**Azure Functions**

The following example demonstrates the same process performed with an AWS Lambda function, but instead utilizing Azure Functions within the Azure environment. This further proves that the method can be employed across various cloud-based environments.

**Conclusion and Defense Summary****Google’s Response**

As described in Tenable’s article, Google responded to their research by creating a remediation patch. This update altered the default behavior of Cloud Build and the default Cloud Build SA. Additionally, new organization policies were released to give organizations full control over which SA Cloud Build uses by default. While Google has implemented this remediation, Cloud Build services can still be used to execute non-privileged commands as a means of enumerating an environment.

**Mitigation Summary**

The most effective mitigation strategy to protect your environment from similar threat actor behavior is to ensure that all SAs within your cloud environment adhere to the principle of least privilege and that no legacy cloud SAs are still in use. Ensure that all cloud services and dependencies are up to date with the latest security patches. If legacy SAs are present, replace them with least-privilege SAs. 

Additionally, users with access to Cloud Functions should not have IAM permissions to the services included in the function’s orchestration.

**Threat Hunting Recommendations**

1.  **Audit and monitor SA permissions**: Regularly audit and monitor SA permissions, with a particular focus on the default Cloud Build SA. Adhere to the principle of least privilege by removing any excessive permissions that are not essential for the SA’s operations.
2.  **Alert setup for Cloud Functions**: Establish alerts for any unusual or unauthorized creation or modification of Cloud Functions. Identify potentially malicious activities where an attacker may be attempting to exploit function deployments for privilege escalation.
3.  **Inspect network traffic**: Analyze network traffic for unusual patterns or connections that might indicate data exfiltration attempts. Pay attention to data being sent to unknown or unauthorized external endpoints, such as those using Ngrok or similar tunneling services.
4.  **Verify NPM package integrity**: Ensure the integrity and authenticity of NPM packages used within Cloud Functions. Prevent the execution of malicious scripts embedded in package.json files that could facilitate environment enumeration or other malicious activities.
5.  **Detect environment enumeration**: Detect and respond to signs of environment enumeration, such as ICMP discovery or system information gathering.

Duping Cloud Functions: An emerging serverless attack vector

In this week’s newsletter, Thor inspects the LockBit leak, finding $10,000 “security tips,” ransom negotiations gone wrong and a rare glimpse into the human side of cybercrime.

Thursday, May 15, 2025 14:01

Welcome to this week’s edition of the Threat Source newsletter. 

I haven’t been to Prague in a while, which is a pity. It’s a wonderful city — great people, amazing food. I’ve visited customers there, held team meetings at the local office (shoutout to Petr!) and spent some memorable summer days off. But none of those are why I’m sending my greetings this time. 

Last week, anyone trying to access LockBit’s dark web affiliate panels was greeted by a defaced page with the message: 

> "Don't do crime CRIME IS BAD xoxo from Prague” 

Alongside the message was a download link for a compressed archive called “paneldb\_dump.zip” —  a 7.5MB file that extracts to a 26MB clear-text SQL dump containing 20 tables. The breach exposed a rare, unfiltered look into LockBit’s operations. 

While most articles focused on the nearly 60,000 Bitcoin addresses or the credentials for 75 admins and affiliates (all with plaintext passwords), I have to admit that I was mesmerized by the “chats” table. 4,423 messages distributed across 208 victims, spanning from Dec. 2024 to April 2025, these chats reveal the raw tactics, ransom demands and negotiation strategies of both affiliates and victims. Sometimes there was just a single unanswered message; in other cases, over 300 messages included “technical support” for unrecoverable files, and even requests for refunds.  

Ransom demands varied widely, from just a few thousand dollars to as much as $2 million in one notable case. There were also several instances of confusion — some mistakenly thought the demand was "100,000 bitcoins" when it was actually "100,000 dollars in bitcoin.” Additionally, there was a case involving a hosting company breach, where it was the company’s customers who ultimately suffered the consequences. The chat exposed that LockBit encrypted all the data with the same key; even though not all victims were willing or able to pay the ransom, LockBit insisted the hoster pay the full amount, making it difficult to collect the asked ransom. 

Negotiations were often pressured by tight deadlines, but European bank holidays on Good Friday, Easter Monday and May 1st further complicated the situation. Multiple times there were situations where the ransom demand increased after a specific deadline. I even found messages from victims asking for more time so they could gather funds in smaller amounts to avoid detection under local anti-money laundering laws. 

In another chat, a victim tried to negotiate by pleading inability to pay a $100k ransom, only to be told, “Seven directors at 14k can’t chime in?” This clearly shows that the “Analytics Department” of LockBit did their homework. 

The level of “trust” placed in affiliates was also striking. Messages included:

Interestingly, that last service was offered for an extra fee. Let me share some of their $10,000 “tips” for free:

With these $10,000 tips, I personally think it would be better to get advice before an incident from Talos Incident Response. They can also provide guidance and proactive support as part of the Talos IR Retainer.

The LockBit leak is a rare window into the mechanics of cybercrime and the human stories behind the headlines. And, for now, xoxo to Prague.

**The one big thing **

Cisco Talos has observed a growing trend of attack kill chains being split into two stages — initial compromise and subsequent exploitation — executed by separate threat actors. In response to these evolving threats, we have refined the definitions of initial access brokers (IABs) to include subcategories such as financially-motivated initial access (FIA), state-sponsored initial access (SIA), and opportunistic initial access (OIA).   

**Why do I care? **

This trend complicates traditional threat modeling and actor profiling, as it requires understanding the intricate relationships and interactions between various groups. For example, hunting and containment strategies that may defend against one type of IAB may not be suitable for another. 

**So now what? **

We have identified several methods for analyzing compartmentalized attacks and propose an extended Diamond Model, which adds a “Relationship Layer” to enrich the context of the relationships between the four features. Familiarize yourself with the new taxonomy we propose, and incorporate this new methodology for modeling and tracking compartmentalized threats into your toolkit. 

**Top security headlines of the week **

**Operation Moonlander**    
A criminal proxy network that has been around for more than 20 years and was built on thousands of infected IOT and end-of-life (EoL) devices was dismantled in an international operation. (U.S. Attorney’s Office) 

**Supply Chain Compromise**    
A deprecated node.js package with more than 40k downloads per week, ‘rand-user-agent' has been compromised with a malicious payload dubbed “RATatouille”. This is a clear case of a supply chain attack. (Aikido) 

**Ascension Health Data Breach Impacts Over 430,000**   
Healthcare provider Ascension has disclosed a data breach affecting over 430,000 patients.  (Bleeping Computer) 

**Germany Shuts Down eXch Over $1.9B Laundering**   
German authorities have shut down the cryptocurrency mixer eXch due to its alleged involvement in laundering approximately $1.9 billion in illicit funds, seizing a large amount of cryptocurrency and data. (BKA, German language)

**Can’t get enough Talos? **

**Talos Takes**  
Follow the motive: Rethinking defense against Initial Access Groups. Listen here.

**Talos in the news**  
Initial Access Brokers Target Brazil Execs via NF-e Spam and Legit RMM Trials (The Hacker News) 

Why MFA is getting easier to bypass and what to do about it (Ars Tecnnica)

**Upcoming events where you can find Talos **

*   Cisco Connect UK & Ireland (May 20) London, UK  
*   BotConf (May 20 – 23) Angers, France  
*   Cisco Live U.S. (June 8 – 12) San Diego, CA

**Most prevalent malware files from Talos telemetry over the past week  **

**SHA 256: e00aa8146cf1202d8ba4fffbcf86da3c6d8148a80bb6503d89b0db2aa9cc0997**   
MD5: eae884415e5fd403e4f1bf46f90df0be   
VirusTotal: https://www.virustotal.com/gui/file/e00aa8146cf1202d8ba4fffbcf86da3c6d8148a80bb6503d89b0db2aa9cc0997    
Typical Filename: paneldb\_dump.zip 

**SHA 256: 9f1f11a708d393e0a4109ae189bc64f1f3e312653dcf317a2bd406f18ffcc507**   
MD5: 2915b3f8b703eb744fc54c81f4a9c67f   
VirusTotal: https://www.virustotal.com/gui/file/9f1f11a708d393e0a4109ae189bc64f1f3e312653dcf317a2bd406f18ffcc507   
Typical Filename: VID001.exe   
Claimed Product: N/A   
Detection Name: Win.Worm.Coinminer::1201 

**SHA 256: a31f222fc283227f5e7988d1ad9c0aecd66d58bb7b4d8518ae23e110308dbf91**   
MD5: 7bdbd180c081fa63ca94f9c22c457376   
VirusTotal: https://www.virustotal.com/gui/file/a31f222fc283227f5e7988d1ad9c0aecd66d58bb7b4d8518ae23e110308dbf91    
Typical Filename: c0dwjdi6a.dll   
Claimed Product: N/A   
Detection Name: Trojan.GenericKD.33515991

Xoxo to Prague

Microsoft has released its monthly security update for May of 2025 which includes 78 vulnerabilities affecting a range of products, including 11 that Microsoft marked as “critical”.  
Microsoft noted five vulnerabilities that have been observed to be exploited in the wild. CVE-2025-30397 is a remote code

Tuesday, May 13, 2025 16:38

Microsoft has released its monthly security update for May of 2025 which includes 78 vulnerabilities affecting a range of products, including 11 that Microsoft marked as “critical”.  

Microsoft noted five vulnerabilities that have been observed to be exploited in the wild. CVE-2025-30397 is a remote code execution vulnerability in the Microsoft Scripting Engine. There were also four elevation of privilege vulnerabilities being actively exploited, CVE-2025-32709, CVE-2025-30400, CVE-2025-32701 and CVE-2025-32706 affecting the Ancillary Function Driver for WinSock, the DWM Core Library and the Windows Common Log File System Driver.  

The eleven "critical” entries consist of five remote code execution (RCE) vulnerabilities, four elevation of privilege vulnerabilities, one information disclosure vulnerability and one spoofing vulnerability. Three of the critical vulnerabilities have been marked as "Exploitation more likely": CVE-2025-30386 --a Microsoft Office RCE vulnerability, CVE-2025-30390 --an Azure ML Compute elevation of privilege vulnerability, and CVE-2025-30398 – a Nuance PowerScribe 360 information disclosure vulnerability.  

The most notable of the “critical” vulnerabilities listed affect Microsoft Office. CVE-2025-30386 is a RCE vulnerability with base CVSS 3.1 score of 8.3. To successfully exploit CVE-2025-30386, an attacker could send a victim an email, and without the victim clicking the link, viewing or interacting with the email, trigger a use-after-free scenario, allowing arbitrary code to be executed. Microsoft has assessed that the attack complexity is “Low”, and exploitation is “More likely”. Another RCE vulnerability affecting Microsoft Office, CVE-2025-30377, has a CVSS 3.1 base score of 8.4, and has been assessed an attack complexity of “Low”, but exploitation is considered “Less Likely”. 

Two RCE vulnerabilities affect the Remote Desktop Client. CVE-2025-29966 and CVE-2025-29967 are both Heap-cased Buffer Overflow vulnerabilities with CVSS 3.1 base scores of 8.8 with “Low” attack complexity and exploitation “Less Likely”. An attacker controlling a Remote Desktop Server could trigger the buffer overflow in a vulnerable when a vulnerable Remote Desktop Client connects to the server. 

CVE-2025-29833 is a RCE affecting the Virtual Machine Bus. This is a Time-of-check Time-of-use (TOCTOU) Race Condition which has been assessed an attack complexity of “High” and exploitation is “Less Likely”. 

Talos would also like to highlight the following "important" vulnerabilities as Microsoft has determined that exploitation is "More likely": 

*   CVE-2025-24063 - Kernel Streaming Service Driver Elevation of Privilege Vulnerability 
*   CVE-2025-29841 - Universal Print Management Service Elevation of Privilege Vulnerability 
*   CVE-2025-29971 - Web Threat Defense (WTD.sys) Denial of Service Vulnerability 
*   CVE-2025-29976 - Microsoft SharePoint Server Elevation of Privilege Vulnerability 
*   CVE-2025-30382 - Microsoft SharePoint Server Remote Code Execution Vulnerability 
*   CVE-2025-30385 - Windows Common Log File System Driver Elevation of Privilege Vulnerability 
*   CVE-2025-30388 - Windows Graphics Component Remote Code Execution Vulnerability

  
A complete list of all the other vulnerabilities Microsoft disclosed this month is available on its update page.  

In response to these vulnerability disclosures, Talos is releasing a new Snort rule set that detects attempts to exploit some of them. Please note that additional rules may be released at a future date and current rules are subject to change pending additional information. Cisco Security Firewall customers should use the latest update to their ruleset by updating their SRU. Open-source Snort Subscriber Rule Set customers can stay up to date by downloading the latest rule pack available for purchase on Snort.org. 

The rules included in this release that protect against the exploitation of many of these vulnerabilities are 64848-64867. There are also these Snort 3 rules: 64852-64853, 301192-301200, and 301203

Microsoft Patch Tuesday for May 2025 — Snort rules and prominent vulnerabilities

How do you profile actors and defend your systems when multiple threat actors are working together? In Part 2, Cisco Talos proposes an extended Diamond Model to analyze complex relationships between attackers.

Defining a new methodology for modeling and tracking compartmentalized threats

Threat actors are teaming up, splitting attacks into stages and making defense harder than ever. In Part 1, Cisco Talos examines their tactics and defines their motivations.

Source

TALOS