Back to Intelligence

WordlistLoader and SynkLoader: Defending Against ClickFix-Delivered Amatera Stealer and Credential Phishing

SA
Security Arsenal Team
August 24, 2026
12 min read

Gen Digital's threat research team has flagged two new loader families — WordlistLoader and SynkLoader — that represent the latest evolution in the initial-access supply chain. WordlistLoader is being distributed through ClearFake campaigns that abuse the ClickFix (aka FakeCaptcha) social-engineering technique to deliver Amatera Stealer (also tracked as ACR Stealer or AcridRain Stealer). SynkLoader, meanwhile, is being used to phish Windows passwords directly from victims.

Why should defenders care? Because loaders like these are not the endgame — they are the front door. The reporting indicates these families are likely used to sell access to encryption-based cyber incident groups — in plain terms, ransomware affiliates and initial access brokers (IABs). A stealer infection that harvests browser credentials, session cookies, and authentication tokens today becomes a ransomware detonation in your environment next quarter. Every SOC needs detections for the ClickFix execution chain and Amatera staging behavior deployed now, not after the first incident ticket.

Technical Analysis

The Threat Actors and Tooling

ComponentRoleNotes
ClearFakeDistribution frameworkMalicious JavaScript injected into compromised websites, displaying fake browser-update or CAPTCHA pages
ClickFix / FakeCaptchaSocial engineering techniqueTricks users into manually copying and pasting a malicious command into the Windows Run dialog, PowerShell, or Terminal — bypassing most email/attachment defenses because the user executes the malware themselves
WordlistLoaderFirst-stage loaderRetrieves and executes next-stage payloads; observed delivering Amatera
Amatera Stealer (ACR / AcridRain)Infostealer payloadHarvests browser credentials, cookies, crypto wallets, and session tokens; sold as malware-as-a-service
SynkLoaderLoader / credential phisherUsed to phish Windows account passwords directly

How the ClickFix Attack Chain Works (Defender's View)

The ClickFix technique is effective precisely because it short-circuits traditional perimeter defenses. There is no malicious attachment to sandbox and no exploit to patch. The chain typically looks like this:

  1. Victim lands on a compromised or malicious site injected with ClearFake JavaScript. The page displays a fake CAPTCHA ("Verify you are human") or a fake browser/Chrome update prompt.
  2. The page copies a malicious command to the victim's clipboard via JavaScript and instructs the user to press Win+R, paste (Ctrl+V), and hit Enter — framed as a "verification step."
  3. The pasted command is typically a mshta.exe, powershell.exe, or wscript.exe one-liner that reaches out to an attacker-controlled host, retrieves the next stage (WordlistLoader), and executes it in memory or from a user-writable directory like %TEMP% or %APPDATA%.
  4. WordlistLoader stages Amatera Stealer, which immediately begins harvesting browser credential stores (Chrome, Edge, Firefox), cookies, autofill data, cryptocurrency wallet files, and session tokens, then exfiltrates over HTTP/S to C2 infrastructure.
  5. SynkLoader operates in parallel campaigns, presenting fraudulent Windows credential dialogs or phishing pages to capture domain/local account passwords — high-value material for IAB resale.

The critical detection opportunity: the user's manual execution means the parent process of the malicious command is explorer.exe (via the Run dialog) — and the command line will almost always contain a URL, an encoded payload, or a LOLBin invocation. That is anomalous in virtually every enterprise environment.

Exploitation Status

  • Active in the wild: Yes — these are observed, operational campaigns, not proof-of-concept research.
  • No CVE is associated with this activity. ClickFix exploits human behavior and legitimate Windows binaries, not a software vulnerability. There is nothing to patch; this must be countered with detection engineering, hardening, and user awareness.
  • Ransomware nexus: The likely sale of access to encryption-focused groups elevates this from "nuisance stealer" to a pre-ransomware indicator. Treat any Amatera/WordlistLoader detection as a potential precursor to a full intrusion and hunt accordingly (check for follow-on discovery, lateral movement, and persistence within 24-72 hours of the initial event).

Detection & Response

SIGMA Rules

The following rules target the highest-fidelity observables in the ClickFix → WordlistLoader → Amatera chain: (1) user-driven LOLBin execution of remote content via the Run dialog, (2) mshta spawning script interpreters, and (3) suspicious scripted downloads into user-writable paths.

YAML
---
title: ClickFix FakeCaptcha User-Executed Remote Command via Run Dialog
id: 3f8a1c92-7d4e-4b5a-9f21-8c6e2a4d7b10
status: experimental
description: Detects the ClickFix/FakeCaptcha social engineering pattern where a user is tricked into pasting a malicious command into the Windows Run dialog, resulting in explorer.exe spawning mshta, powershell, or wscript with a URL or encoded payload. Associated with ClearFake campaigns delivering WordlistLoader and Amatera Stealer.
references:
  - https://thehackernews.com/2026/08/wordlistloader-delivers-amatera-via.html
  - https://attack.mitre.org/techniques/T1204/001/
  - https://attack.mitre.org/techniques/T1218/005/
author: Security Arsenal
date: 2026/08/15
tags:
  - attack.execution
  - attack.t1204.001
  - attack.t1218.005
  - attack.t1059.001
logsource:
  category: process_creation
  product: windows
detection:
  selection_parent:
    ParentImage|endswith: '\explorer.exe'
  selection_child:
    Image|endswith:
      - '\mshta.exe'
      - '\powershell.exe'
      - '\pwsh.exe'
      - '\wscript.exe'
      - '\cscript.exe'
      - '\curl.exe'
  selection_cli:
    CommandLine|contains:
      - 'http://'
      - 'https://'
      - '-enc'
      - '-e '
      - 'FromBase64String'
      - 'IEX'
      - 'Invoke-Expression'
      - 'iwr '
      - 'Invoke-WebRequest'
      - 'DownloadString'
      - 'mshta vbscript'
  condition: selection_parent and selection_child and selection_cli
falsepositives:
  - Rare administrative troubleshooting where an admin pastes commands into the Run dialog
level: high
---
title: Mshta Spawning Script Interpreter or Downloader Child Process
id: 6b2e4d18-9a3c-4f7b-8e15-2d9c5a7e3f06
status: experimental
description: Detects mshta.exe spawning PowerShell, cmd, or other script interpreters — a common ClickFix and loader staging pattern observed in WordlistLoader delivery of Amatera Stealer via ClearFake campaigns.
references:
  - https://thehackernews.com/2026/08/wordlistloader-delivers-amatera-via.html
  - https://attack.mitre.org/techniques/T1218/005/
author: Security Arsenal
date: 2026/08/15
tags:
  - attack.defense_evasion
  - attack.execution
  - attack.t1218.005
logsource:
  category: process_creation
  product: windows
detection:
  selection:
    ParentImage|endswith: '\mshta.exe'
    Image|endswith:
      - '\powershell.exe'
      - '\pwsh.exe'
      - '\cmd.exe'
      - '\wscript.exe'
      - '\cscript.exe'
      - '\rundll32.exe'
      - '\regsvr32.exe'
  condition: selection
falsepositives:
  - Legacy line-of-business HTA applications (rare in modern environments; baseline and allowlist)
level: high
---
title: Suspicious Payload Staging in User-Writable Directory Followed by Execution
id: 9d1c7f34-2b8e-4a6d-b392-5e8f4c1a9d27
status: experimental
description: Detects creation and subsequent execution of script or executable payloads in TEMP/APPDATA user-writable directories, consistent with WordlistLoader/Amatera Stealer staging behavior delivered through ClickFix lures.
references:
  - https://thehackernews.com/2026/08/wordlistloader-delivers-amatera-via.html
  - https://attack.mitre.org/techniques/T1059/
  - https://attack.mitre.org/techniques/T1036/
author: Security Arsenal
date: 2026/08/15
tags:
  - attack.execution
  - attack.t1059
  - attack.t1036
logsource:
  category: process_creation
  product: windows
detection:
  selection_path:
    Image|contains:
      - '\AppData\Local\Temp\'
      - '\AppData\Roaming\'
      - '\Users\Public\'
      - '\ProgramData\'
  selection_ext:
    Image|endswith:
      - '.exe'
      - '.ps1'
      - '.hta'
      - '.js'
      - '.vbs'
      - '.bat'
  filter_known:
    Image|contains:
      - '\AppData\Local\Microsoft\'
      - '\AppData\Local\Google\'
      - '\AppData\Roaming\Microsoft\'
      - '\AppData\Local\Temp\7z'
      - '\AppData\Local\Temp\Rar$'
      - '\AppData\Local\Temp\nsg'
  condition: selection_path and selection_ext and not filter_known
falsepositives:
  - Software updaters (Teams, Zoom, OneDrive) installing per-user — tune with additional signer/parent filters for your environment
level: medium

KQL Hunt (Microsoft Sentinel / Defender)

This query hunts the ClickFix execution pattern across both Defender telemetry (DeviceProcessEvents) and Sysmon-forwarded events (SecurityEvent), looking for explorer-spawned LOLBins reaching for remote content — the fingerprint of a user who just pasted a fake-CAPTCHA command:

KQL — Microsoft Sentinel / Defender
let lookback = 14d;
let suspiciousBins = @"mshta.exe|powershell.exe|pwsh.exe|wscript.exe|cscript.exe|curl.exe|bitsadmin.exe|rundll32.exe";
let suspiciousCli = @"http://|https://|-enc|-e |FromBase64String|IEX|Invoke-Expression|Invoke-WebRequest|iwr |DownloadString|hidden|-w hidden|mshta vbscript";
union isfuzzy=true
(DeviceProcessEvents
 | where TimeGenerated > ago(lookback)
 | where InitiatingProcessFileName =~ "explorer.exe"
 | where FileName matches regex suspiciousBins
 | where ProcessCommandLine matches regex suspiciousCli
 | project TimeGenerated, DeviceName, AccountName, FileName, ProcessCommandLine, InitiatingProcessFileName, SHA256, ReportId
 | extend Source = "Defender"),
(SecurityEvent
 | where TimeGenerated > ago(lookback)
 | where EventID == 4688
 | where ParentProcessName endswith @"\explorer.exe"
 | where NewProcessName matches regex suspiciousBins
 | where CommandLine matches regex suspiciousCli
 | project TimeGenerated, Computer, Account, NewProcessName, CommandLine, ParentProcessName, Source = "Sysmon4688");

Pair that with a network-focused hunt for newly seen outbound connections from script interpreters and mshta — Amatera's C2 exfiltration will surface here:

KQL — Microsoft Sentinel / Defender
DeviceNetworkEvents
| where TimeGenerated > ago(7d)
| where InitiatingProcessFileName in~ ("mshta.exe","powershell.exe","wscript.exe","cscript.exe","rundll32.exe")
| where RemoteUrl !has_any ("microsoft.com","windowsupdate.com","office.com","office365.com")  // tune for your environment
| summarize FirstSeen = min(TimeGenerated), LastSeen = max(TimeGenerated), ConnectionCount = count(), Devices = dcount(DeviceName) by InitiatingProcessFileName, RemoteUrl, RemoteIP
| where ConnectionCount < 50   // focus on low-prevalence, novel destinations
| order by FirstSeen desc;

Velociraptor VQL Hunt

For DFIR teams running Velociraptor, this artifact sweeps the fleet for the ClickFix execution pattern — explorer-spawned LOLBins with URLs or encoded content in the command line — plus staged payloads in user-writable paths:

VQL — Velociraptor
-- Hunt for ClickFix/FakeCaptcha execution chains and loader staging
-- Targets explorer-spawned LOLBins with remote URLs or encoded payloads,
-- plus executables staged in user-writable directories (WordlistLoader/Amatera behavior)

LET procs = SELECT Pid, Ppid, Name, Exe, CommandLine, Username, CreateTime
FROM pslist()
WHERE Name =~ '(?i)(mshta|powershell|pwsh|wscript|cscript|curl|bitsadmin)'
  AND (CommandLine =~ '(?i)(https?://|FromBase64String|Invoke-Expression|DownloadString|-enc|hidden)')

LET staged = SELECT FullPath, Size, Mtime, Btime
FROM glob(globs=['C:/Users/*/AppData/Local/Temp/*.exe','C:/Users/*/AppData/Local/Temp/*.ps1','C:/Users/*/AppData/Roaming/*/*.exe','C:/Users/Public/*.exe','C:/Users/Public/*.hta'])
WHERE Mtime > now() - 604800   -- last 7 days

SELECT * FROM procs
UNION ALL
SELECT NULL AS Pid, NULL AS Ppid, 'STAGED_FILE' AS Name, FullPath AS Exe, FullPath AS CommandLine, NULL AS Username, Mtime AS CreateTime FROM staged

Hardening & Verification Script

There is no patch for social engineering — but you can raise the cost dramatically. This PowerShell script (1) audits whether critical attack-surface controls are in place, (2) blocks mshta and script interpreters from launching via the Run-dialog path using Windows Defender Attack Surface Reduction rules and WDAC-adjacent hardening, and (3) verifies PowerShell logging is enabled for detection coverage:

PowerShell
# ============================================================
# Security Arsenal - ClickFix / Loader Hardening & Audit Script
# Run as Administrator. Audit first, then apply.
# ============================================================

# --- 1. AUDIT: Check ASR rule state for known loader-abused rules ---
# ASR rule GUIDs relevant to this threat:
#   d1e49aac-8f56-4280-b9ba-993a6d77406c = Block process creations from PSExec/WMI (context)
#   5beb7efe-fd9a-4556-801d-275e5ffc04cc = Block execution of potentially obfuscated scripts
#   92e97fa1-2edf-4476-bdd6-9dd0b4dddc7b = Block Win32 API calls from Office macros (context)
#   d4f940ab-401b-4efc-aadc-ad5f3c50688a = Block Office apps creating child processes (context)
#   26190899-1602-49e8-8b27-eb1d0a1ce869 = Block Office communication apps creating child processes
$asrRules = @{
  'Obfuscated scripts'      = '5beb7efe-fd9a-4556-801d-275e5ffc04cc'
  'PSExec/WMI child procs'  = 'd1e49aac-8f56-4280-b9ba-993a6d77406c'
}
Write-Host "=== ASR Rule Status ===" -ForegroundColor Cyan
foreach ($name in $asrRules.Keys) {
  $guid = $asrRules[$name]
  $val  = (Get-MpPreference).AttackSurfaceReductionRules_Ids -contains $guid
  $idx  = (Get-MpPreference).AttackSurfaceReductionRules_Ids.IndexOf($guid)
  if ($val) {
    $action = (Get-MpPreference).AttackSurfaceReductionRules_Actions[$idx]
    Write-Host "[$(if($action -eq 1){'BLOCK'}elseif($action -eq 2){'AUDIT'}else{'OTHER'})] $name"
  } else {
    Write-Host "[NOT SET ] $name" -ForegroundColor Yellow
  }
}

# --- 2. HARDEN: Enable ASR rule against obfuscated scripts (set to Audit first, then Block) ---
# Start with 2 (Audit) for 7-14 days, review Defender event 1121/1122, then flip to 1 (Block)
Add-MpPreference -AttackSurfaceReductionRules_Ids 5beb7efe-fd9a-4556-801d-275e5ffc04cc `n  -AttackSurfaceReductionRules_Actions 1

# --- 3. HARDEN: Restrict mshta.exe via Software Restriction-style AppLocker rule check ---
$applocker = Get-AppLockerPolicy -Effective -ErrorAction SilentlyContinue
if (-not $applocker) {
  Write-Host "[!] No AppLocker policy detected. Deploy an AppLocker/WDAC rule to DENY mshta.exe for standard users." -ForegroundColor Yellow
} else {
  Write-Host "[+] AppLocker policy present. Verify mshta.exe is denied for non-admin users."
}

# --- 4. DETECTION: Ensure PowerShell Script Block Logging + Module Logging are enabled ---
$sbPath = 'HKLM:\SOFTWARE\Policies\Microsoft\Windows\PowerShell\ScriptBlockLogging'
if (-not (Test-Path $sbPath)) { New-Item -Path $sbPath -Force | Out-Null }
Set-ItemProperty -Path $sbPath -Name 'EnableScriptBlockLogging' -Value 1
$modPath = 'HKLM:\SOFTWARE\Policies\Microsoft\Windows\PowerShell\ModuleLogging'
if (-not (Test-Path $modPath)) { New-Item -Path $modPath -Force | Out-Null }
Set-ItemProperty -Path $modPath -Name 'EnableModuleLogging' -Value 1
Write-Host "[+] PowerShell Script Block and Module logging enabled."

# --- 5. DETECTION: Verify process creation command-line auditing (Event 4688) ---
$cmdLine = (Get-ItemProperty 'HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System\Audit' `n  -Name 'ProcessCreationIncludeCmdLine_Enabled' -ErrorAction SilentlyContinue).ProcessCreationIncludeCmdLine_Enabled
if ($cmdLine -ne 1) {
  Set-ItemProperty 'HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System\Audit' `n    -Name 'ProcessCreationIncludeCmdLine_Enabled' -Value 1
  Write-Host "[+] Enabled command-line capture for Event 4688."
} else {
  Write-Host "[+] Event 4688 command-line capture already enabled."
}

# --- 6. VERIFY: Defender real-time protection + cloud protection ---
$mp = Get-MpComputerStatus
Write-Host "=== Defender Status ===" -ForegroundColor Cyan
Write-Host "RealTimeProtection: $($mp.RealTimeProtectionEnabled) | CloudProtection: $($mp.IsCloudProtectionEnabled) | SignatureAge(days): $($mp.AntivirusSignatureAge)"
if ($mp.AntivirusSignatureAge -gt 2) { Write-Host "[!] Signatures stale - force update now." -ForegroundColor Red }

Write-Host "`nDone. Review ASR audit events (Defender EventIDs 1121/1122) before enforcing Block mode." -ForegroundColor Green

Remediation

Because this campaign exploits user behavior and legitimate Windows binaries rather than a software vulnerability, "remediation" means hardening, detection, and response readiness:

  1. Block or constrain mshta.exe for standard users. There is virtually no legitimate reason for a standard user to execute mshta in a modern enterprise. Use AppLocker or WDAC to deny it outright, or at minimum alert on every execution.
  2. Deploy the Sigma/KQL detections above in audit mode first, tune for your software-updater noise, then move to alerting with high priority. The explorer-parent + URL-in-command-line pattern is a genuine high-fidelity signal.
  3. Enable PowerShell Script Block Logging and Module Logging (Event 4104/4103) and forward to your SIEM — encoded and staged ClickFix payloads are highly visible there even when command lines are partially obfuscated.
  4. Enable Defender ASR rules (start with obfuscated-script blocking in audit mode; review EventIDs 1121/1122; enforce block after tuning).
  5. User awareness, but specific. Generic phishing training won't stop ClickFix. Show users the actual lure: "No legitimate website will ever ask you to press Win+R and paste a command to prove you're human or to fix a browser error." Make that one sentence part of onboarding and quarterly training.
  6. Treat stealer detections as pre-ransomware. If Amatera/ACR Stealer is confirmed on any host: isolate the endpoint, force enterprise-wide credential resets for every account that touched that machine (including browser-saved credentials and session tokens — revoke active sessions/OAuth grants, not just passwords), rotate any service-account secrets stored on the host, and hunt 72 hours backward and forward for follow-on activity before the host is reimaged.
  7. Browser credential hygiene. Move users to an enterprise password manager with credentials not stored in the browser, enforce MFA everywhere (stealers thrive on password reuse), and consider hardware-backed (FIDO2) authentication for privileged and remote-access accounts — stolen cookies are how these crews bypass MFA, so enforce short session lifetimes and conditional access on token-theft indicators.
  8. Web filtering and script control. Block newly registered and uncategorized domains at the proxy, and consider blocking .hta file association and HTA MIME delivery at the perimeter.

Incident Response Pointers

  • Contain: isolate host, disable the user account in IdP, revoke sessions.
  • Investigate: pull the Run dialog execution chain (the Sigma/KQL queries above give you the pivot), check clipboard-monitoring telemetry if your EDR supports it, and identify the lure domain for threat-intel reporting and blocklisting.
  • Report: ClearFake/ClickFix infrastructure changes rapidly; share IoCs with your ISAC and upstream providers.

The bottom line: WordlistLoader and SynkLoader are the delivery mechanism for the next incident — the ransomware deployment that follows the access sale. Detecting and killing this chain at the ClickFix execution step is the cheapest, highest-leverage control point you have.

Related Resources

Security Arsenal Healthcare Cybersecurity AlertMonitor Platform Book a SOC Assessment healthcare Intel Hub

Is your security operations ready?

Get a free SOC assessment or see how AlertMonitor cuts through alert noise with automated triage.