Threat reportMalwareTL-2026-3328

Casbaneiro Banking Trojan Activates Only When Victims Visit Targeted Bank Websites

highACTIVE

Casbaneiro Banking Trojan Activates Only When Victims Visit (TL-2026-3328), also tracked as Casbaneiro, is a high-severity malware campaign, first published 2026-10-12. It has no confirmed attribution, affects Microsoft Windows (including Outlook address-book data on infected, maps to 21 MITRE ATT&CK techniques (T1027, T1047, T1048.003), and is covered by 9 detection rules and 45 indicators of compromise.

Severity
HIGHAssessed severity
CVEs
0None referenced
Techniques
21MITRE ATT&CK
Actors
0Not attributed
Detection rules
9SPL · KQL · Sigma
IOCs
45Indicators of compromise

Key facts for TL-2026-3328

Threat ID
TL-2026-3328
Also known as
Casbaneiro
Severity
HIGH
Status
ACTIVE
Category
MALWARE
First published
Last reviewed
Attribution confidence
LOW
Motivation
FINANCIAL
Target sectors
finance, banking, general consumers
Target regions
argentina, peru, colombia, mexico, Latin America
Detection rules
9
Indicators of compromise
45
Updates
2026-10-12 · revalidated 1× · latest source

Malware and tooling in Casbaneiro Banking Trojan Activates Only When Victims Visit

Malware and tooling: Metamorfo, Rakhni

How Casbaneiro Banking Trojan Activates Only When Victims Visit works

A Casbaneiro banking trojan campaign observed in August 2026 targets users in Argentina, Peru, Colombia and Mexico through phishing emails with PDFs posing as invoices or legal notices. A geofenced chain (ZIP, HTA, JScript, AutoIt loader) injects the payload into RegSvcs.exe or mobsync.exe, and banking C2 activity starts only when the victim visits a targeted bank website.

FortiGuard Labs documented a Casbaneiro campaign observed in August 2026 and published on 2026-09-10; GBHackers re-reported it on 2026-10-12. Initial access is a phishing email or PDF disguised as an invoice or legal notice, personalized with the recipient's email address and country-code domains. The links lead to a geofenced web page: visitors outside the target regions are redirected to legitimate sites such as Google or YouTube, while eligible visitors receive a ZIP archive delivered through Base64-encoded content embedded in JavaScript.

The ZIP contains an HTA downloader that references external JavaScript. That script retrieves an XML-based script package containing JScript, which performs WMI-based environment checks (sandbox detection and OS language identification) and proceeds only if the OS language matches whitelisted countries. The loader then downloads, as separate components to hinder static detection, a legitimate AutoIt interpreter, a compiled AutoIt script and a compressed payload into a randomly named directory (C:\{random name}). The AutoIt loader displays a fake Windows service window, decompresses the payload and injects Casbaneiro into RegSvcs.exe, falling back to mobsync.exe.

The payload decrypts cryptocurrency addresses, a global ID and exfiltration URLs using fragmented decryption keys. It creates the mutex GlobolID-4465173{Username}, exits if the default system language is German, French or English, and builds an MD5 hash from the computer name, username and executable name. It harvests Outlook address-book entries and sender/recipient information and sends them unencrypted to exfiltration URLs, leaving a .Outlook marker in %APPDATA%. Victim information is sent Base64-encoded to a separate server that is expected to return HTTP 403 Forbidden; any other status causes a retry. Only after a 403 does it create an MD5-named folder in %TEMP% and an MD5-named registry key under HKCU\SOFTWARE. Persistence is an LNK file in the Startup folder that launches the AutoIt script through the interpreter, plus a {ComputerName}@4{UserName} folder in %PUBLIC% as an infection marker.

Banking C2 communication is triggered only when the victim visits a targeted financial website in a browser. Supported tasks are keyboard control, clipboard pasting, file execution, command execution and bank-themed fake windows to facilitate fraud. C2 traffic uses malformed HTTP requests (missing Host header, oversized Content-Length, payload fragmented across small packets), and the distributed servers return misleading 403 responses. The string-decryption algorithm overlaps with the Ousaban banking trojan; the source describes this as an overlap of evasion patterns, not shared operators. No threat actor is named. FortiGuard detection names: PDF/Phishing.5BB0!tr, JS/Phishing.IBP!tr and W32/Casbaneiro.EN!tr.spy.

MITRE ATT&CK techniques used in TL-2026-3328

Defense Evasion

T1027 Obfuscated Files or Information; T1055 Process Injection; T1140 Deobfuscate/Decode Files or Information; T1218.005 Mshta; T1480 Execution Guardrails; T1497.001 System Checks

Execution

T1047 Windows Management Instrumentation; T1059.007 JavaScript; T1059.010 AutoHotKey & AutoIT; T1204.001 User Execution; T1204.002 Malicious File

Exfiltration

T1048.003 Exfiltration Over Unencrypted Non-C2 Protocol

Command and Control

T1071.001 Web Protocols; T1105 Ingress Tool Transfer; T1132.001 Standard Encoding

Collection

T1114.001 Local Email Collection; T1115 Clipboard Data

Persistence

T1547.001 Registry Run Keys / Startup Folder

Initial Access

T1566.001 Spearphishing Attachment; T1566.002 Spearphishing Link

Discovery

T1614.001 System Language Discovery

Affected products and versions in Casbaneiro Banking Trojan Activates Only When Victims Visit

  • Microsoft — Windows (including Outlook address-book data on infected hosts)
    Vulnerable versions: Not specified in source

Remediation for Casbaneiro Banking Trojan Activates Only When Victims Visit

Immediate actions

  • Block the listed domains, IPs and file hashes at email, proxy and endpoint layers
  • Hunt for mshta.exe or wscript/cscript spawning AutoIt interpreters from randomly named directories under C:\
  • Hunt for RegSvcs.exe or mobsync.exe running with AutoIt-related command lines or with unexpected network activity
  • Search for the mutex GlobolID-4465173*, a .Outlook file in %APPDATA%, and {ComputerName}@4{UserName} folders in %PUBLIC%
  • Review Startup-folder LNK files that launch an AutoIt interpreter and script

Workarounds

  • Restrict mshta.exe execution where it is not needed for business operations
  • Block execution of AutoIt interpreters from user-writable or non-standard directories

Longer-term hardening

  • Use email filtering and CDR (Content Disarm and Reconstruction) to neutralize malicious PDF content
  • Keep endpoint protection updated with current Casbaneiro signatures
  • Monitor for unusual HTA and AutoIt execution patterns that indicate early-stage infection
  • Alert on HTTP requests with a missing Host header or oversized Content-Length from endpoints

Timeline of Casbaneiro Banking Trojan Activates Only When Victims Visit

  • FortiGuard Labs observed the Casbaneiro phishing campaign targeting Argentina, Peru, Colombia and Mexico (month-level date: August 2026).
  • FortiGuard noted Casbaneiro uses the same string-decryption algorithm as Ousaban, with the key and encrypted strings split into fragments concatenated at runtime; described as overlapping evasion patterns, not shared operators.
  • FortiGuard named detections PDF/Phishing.5BB0!tr, JS/Phishing.IBP!tr and W32/Casbaneiro.EN!tr.spy for the phishing PDFs, JS stage and Casbaneiro payload.
  • The FortiGuard report released IOCs: PDF, email, HTA, AutoIt script and payload hashes, secureserver.net-hosted domains, gexwalltool.com, x-wolverine.servebbs.com, two IPs and two cryptocurrency addresses.
  • FortiGuard Labs published 'Casbaneiro: A Banking Trojan with Distributed Data Receiving Servers', detailing the geofenced HTA/JScript/AutoIt chain and injection into RegSvcs.exe/mobsync.exe.
  • SOC Prime published an active-threat entry with detection content for the Casbaneiro campaign (MSHTA misuse, RegSvcs/mobsync injection, HKCU\SOFTWARE MD5 keys, HTTP 403 distributed C2).
  • GBHackers published coverage highlighting that C2 banking activity starts only when victims visit targeted bank websites.

Update history for TL-2026-3328

Sources cited for Casbaneiro Banking Trojan Activates Only When Victims Visit

Detection coverage for TL-2026-3328

As of 2026-10-12, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-3328 across Splunk SPL, Microsoft KQL and Sigma, covering 45 indicator(s) of compromise. The whole corpus is readable without an account; a free account unlocks full detection query text in Splunk SPL, Microsoft KQL and Sigma; paid tiers add raw indicator values, correlation and the MCP server. Threadlinqs MCP server · View plans.

9 detection rules (Splunk SPL, Microsoft KQL, Sigma) · Blue and above. Compare plans
45 indicators of compromise · Red and above. Compare plans

Threadlinqs Intelligence — Real-Time Threat Detection Platform

[ 0 threats ] [ 0 det ] [ CRIT: 0 ] [ HIGH: 0 ]
// threat_feed
$ sort --newest
Showing all threats

Live intelligence console

Threat level
Fig. 01 · Threat weatherIndexing the archive…
1 square = 1 threat · click to open

Latest Threats