NETWIRE

Invicta Stealer Spreads Via Fake GoDaddy Refund Invoices

Invicta Stealer is being spread through phishing emails that impersonate GoDaddy refund invoices, using an infection chain that begins with a malicious HTML attachment and progresses through ZIP, LNK, HTA, and PowerShell stages before installing the information-stealing malware. Researchers said the malware’s developer has promoted the family on Telegram, YouTube, and GitHub, including a free builder that appears to have lowered the barrier to entry for other threat actors and increased the stealer’s circulation. Once executed by the victim, Invicta Stealer gathers extensive host and user data, including browser information, Discord data, cryptocurrency wallet contents, Steam and KeePass artifacts, installed application details, and files from Desktop and Documents, then compresses and exfiltrates the data to a Discord webhook or other command-and-control endpoint. The campaign aligns with common user execution tradecraft in phishing-led intrusions, and the malware also uses anti-analysis features such as encrypted strings, syscalls, and multithreading while collecting system and application context from infected machines.

Detect Credential Access with Elastic Security | Elastic Security Labs

Elastic Endpoint Security added kernel-driver-derived file and registry telemetry to detect access to sensitive credential stores, including browser data, Windows SAM and LSA secrets, cached domain credentials, credential files, and Windows Credential Manager. Its behavior protections can block high-confidence credential-theft activity associated with tools and malware such as Mimikatz, LaZagne, AgentTesla, FormBook, and Poulight Stealer; KQL and EQL hunting can also identify processes opening multiple sensitive stores or SMB access following a network logon. Attackers commonly extract saved browser usernames, passwords, cookies, and session tokens from stores such as Chrome’s Login Data and Firefox’s key3.db, key4.db, and logins.json. They also dump Windows LSA secrets and cached domain credentials using utilities including Mimikatz, reg save, Impacket secretsdump.py, CrackMapExec, and gsecdump. Cached Windows credentials are generally stored as DCC2/MS-Cache v2 hashes and require offline cracking rather than pass-the-hash, while Linux Active Directory integrations may retain cached credentials in SSSD or Quest database files.

Cyble - BianLian: New Ransomware Variant On The Rise

BianLian emerged as a Go-based ransomware family targeting organizations in manufacturing, education, healthcare, and BFSI, with victims reporting file encryption and double-extortion threats. The malware appends the .bianlian extension to encrypted files, drops a ransom note named "Look at this instruction.txt", and warns that allegedly stolen financial, client, business, technical, and personal data will be leaked within ten days unless payment is made through negotiations conducted over TOX Messenger and a Tor-based leak site. Technical analysis shows the malware was built to hinder detection and speed impact on victim systems. BianLian performs anti-analysis checks associated with MITRE ATT&CK T1497 virtualization and sandbox evasion, creates multiple threads to accelerate encryption, enumerates drives from A: through Z:, excludes selected files and folders from encryption, and deletes itself after execution. Its behavior also aligns with broader defense-evasion patterns such as executable obfuscation and packing captured in MITRE ATT&CK T1027.002, underscoring how the ransomware combines rapid file encryption with techniques intended to frustrate automated analysis and reverse engineering.

Virus Bulletin :: Collector-stealer: a Russian origin credential and information extractor

Collector Stealer—also marketed as COLLECTOR Project, CollectorGoomba, and formerly Memory Project—was sold as a Russian-language spyware service and used to steal saved browser credentials, cookies, personal data, screenshots, Telegram and Steam data, and cryptocurrency wallet information from infected Windows systems. Researchers said the malware was spread through phishing portals, fake software downloads, and bundled crack or riskware tools such as KMSAuto, often disguised as miners, game utilities, or activation packages. On infected hosts, it gathered data from browsers and applications, captured screenshots, scanned directories, extracted SQLite-stored information, and staged the loot in ZIP or RAR archives before sending it to attacker-controlled panels over HTTP POST. Analysis of the malware’s infrastructure showed that some builds fetched their command-and-control destination from a text file hosted on GitHub, with a fallback to a hard-coded justns.ru subdomain if retrieval failed. After the malicious GitHub repository was reported and removed, affected samples attempted to exfiltrate to an invalid 404: Not Found.ru destination and crashed, temporarily disrupting those variants. Operators later updated newer samples to use upaste[.]me for C2 redirection instead, indicating the stealer remained under active development and continued to evolve its exfiltration workflow and delivery ecosystem.

Cutting-off the Command-and-Control Infrastructure of CollectorGoomba | Threat Bulletin | VMRay