Threat reportAPTTL-2026-6549
AhnLab ASEC: October 2026 APT Attack Trends (Domestic, Korea) - LNK/HWP spear phishing delivering XenoRAT, Xctdoor and .NET RAT
AhnLab ASEC reports APT spear-phishing campaigns against Korean targets observed in October 2026, using malicious LNK and HWP lures, with LNK the most common vector. Payloads include XenoRAT, Xctdoor, an unnamed .NET RAT, PowerShell/AutoIt backdoors and an abused MeshAgent, with Dropbox, GitHub, GitLab and PubNub used for staging and C2.
Key facts
| Identifiers | TL-2026-6549 |
|---|---|
| Severity | HIGH |
| First disclosed | |
| Affected products | Microsoft Windows (endpoints receiving LNK/HWP spear-phishing lures); Hancom Hangul (HWP) documents used as lure format; no vulnerability cited |
| Malware | .NET-based RAT (unnamed), Python, Rakhni, Xctdoor, XenoRAT |
| Tools | AutoIt, MeshAgent |
| MITRE ATT&CK techniques | T1036.004 Masquerade Task or Service, T1053.005 Scheduled Task, T1055 Process Injection, T1056.001 Keylogging, T1057 Process Discovery, T1059.001 PowerShell, T1059.005 Visual Basic, T1059.006 Python, T1082 System Information Discovery, T1113 Screen Capture, T1140 Deobfuscate/Decode Files or Information, T1204.002 Malicious File and 8 more in the attack flow |
| Motivation | UNKNOWN |
| Target sectors | government administration, defense, enterprise, general-users |
| Target regions | south korea |
| Detection rules | 9 (1 shown free) |
| Sources | 7 cited sources |
Detection coverage
As of 2026-11-20, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-6549 across Splunk SPL, Microsoft KQL and Sigma, covering 23 indicator(s) of compromise. The whole corpus is readable without an account. Every threat comes with one complete detection in Splunk SPL, Microsoft KQL and Sigma, free for everyone; a free account adds daily TL Creds to unlock the rest. Paid plans add a larger daily allowance, correlation and the MCP server. Threadlinqs MCP server · View plans.
Free detection: LNK/HWP Spear-Phishing Script Host Chain (mshta, wscript, PowerShell Staging) — Exploitation Phase
Detects the first execution stage of the October 2026 Korean-targeted APT campaign (ASEC, TL-2026-6549): a malicious LNK or HWP lure launching mshta with a remote script, wscript/cscript from user-writable paths, or PowerShell downloading follow-on content from Dropbox, GitLab, GitHub, PubNub or the campaign IPs 104.253.175.177 / 31.59.138.144. Covers Types B, C, D and F delivering XenoRAT, Xctdoor and AutoIt/Python backdoors.
- Languages: Splunk SPL, Microsoft KQL, Sigma
- ATT&CK: T1566.001, T1204.002, T1059.001, T1059.005, T1218.005
- Data sources: sysmon, windows_security
- Severity: high
- Confidence: high
Splunk SPL
index=sysmon sourcetype="XmlWinEventLog:Microsoft-Windows-Sysmon/Operational" EventCode=1
(
(Image="*\\mshta.exe" (CommandLine="*http://*" OR CommandLine="*https://*" OR CommandLine="*vbscript:*"))
OR ((Image="*\\wscript.exe" OR Image="*\\cscript.exe") (CommandLine="*\\AppData\\*" OR CommandLine="*\\Temp\\*" OR CommandLine="*\\Users\\Public\\*" OR CommandLine="*\\ProgramData\\*"))
OR ((Image="*\\powershell.exe" OR Image="*\\pwsh.exe") (CommandLine="*DownloadString*" OR CommandLine="*DownloadFile*" OR CommandLine="*Net.WebClient*" OR CommandLine="*Invoke-WebRequest*" OR CommandLine="*Invoke-RestMethod*" OR CommandLine="*Start-BitsTransfer*" OR CommandLine="*iwr *") (CommandLine="*dropbox*" OR CommandLine="*gitlab*" OR CommandLine="*githubusercontent*" OR CommandLine="*github.com*" OR CommandLine="*pubnub*" OR CommandLine="*104.253.175.177*" OR CommandLine="*31.59.138.144*"))
)
```TL-2026-6549 T1566.001 T1204.002 LNK and HWP lure script-host chain: mshta remote script, wscript/cscript from user-writable paths, PowerShell staging download from Dropbox/GitLab/GitHub/PubNub or campaign IPs```
| eval stage=case(match(Image,"(?i)mshta\\.exe$"),"mshta_remote_script",match(Image,"(?i)(wscript|cscript)\\.exe$"),"script_host_user_path",true(),"powershell_staging_download")
| fillnull value="n/a" User
| stats count min(_time) as first_seen max(_time) as last_seen values(ParentImage) as parent_images values(CommandLine) as command_lines by host, User, stage
| convert ctime(first_seen) ctime(last_seen)
| table first_seen last_seen host User stage parent_images command_lines countMicrosoft KQL
DeviceProcessEvents
| where Timestamp > ago(24h)
| where (FileName =~ "mshta.exe" and ProcessCommandLine has_any ("http://", "https://", "vbscript:"))
or (FileName in~ ("wscript.exe", "cscript.exe") and ProcessCommandLine contains_any (@"\AppData\", @"\Temp\", @"\Users\Public\", @"\ProgramData\"))
or (FileName in~ ("powershell.exe", "pwsh.exe")
and ProcessCommandLine has_any ("DownloadString", "DownloadFile", "Net.WebClient", "Invoke-WebRequest", "Invoke-RestMethod", "Start-BitsTransfer")
and ProcessCommandLine has_any ("dropbox", "gitlab", "githubusercontent", "github.com", "pubnub", "104.253.175.177", "31.59.138.144"))
| extend Stage = case(FileName =~ "mshta.exe", "mshta_remote_script", FileName in~ ("wscript.exe", "cscript.exe"), "script_host_user_path", "powershell_staging_download")
| project Timestamp, DeviceName, AccountName, Stage, FileName, ProcessCommandLine, InitiatingProcessFileName, InitiatingProcessCommandLine, SHA256
| order by Timestamp descSigma
title: LNK/HWP Lure Script Host Chain - Mshta Remote Script, User-Path WScript, PowerShell Staging Download
id: 3f1c2a7e-8b4d-4e61-9a05-6c2d7e1b4a10
status: experimental
description: Detects script-host execution patterns used by the Korean-targeted LNK/HWP spear-phishing campaign in the ASEC October 2026 report (TL-2026-6549), namely mshta loading a remote script, wscript/cscript running from user-writable paths, and PowerShell downloading from Dropbox/GitLab/GitHub/PubNub or the campaign IPs.
references:
- https://asec.ahnlab.com/ko/96683/
author: Threadlinqs Team
date: 2026-11-20
tags:
- attack.initial-access
- attack.t1566.001
- attack.execution
- attack.t1204.002
- attack.t1059.001
- attack.t1059.005
- attack.defense-evasion
- attack.t1218.005
logsource:
category: process_creation
product: windows
detection:
selection_mshta:
Image|endswith: '\mshta.exe'
CommandLine|contains:
- 'http://'
- 'https://'
- 'vbscript:'
selection_wscript:
Image|endswith:
- '\wscript.exe'
- '\cscript.exe'
CommandLine|contains:
- '\AppData\'
- '\Temp\'
- '\Users\Public\'
- '\ProgramData\'
selection_ps:
Image|endswith:
- '\powershell.exe'
- '\pwsh.exe'
selection_ps_download:
CommandLine|contains:
- 'DownloadString'
- 'DownloadFile'
- 'Net.WebClient'
- 'Invoke-WebRequest'
- 'Invoke-RestMethod'
- 'Start-BitsTransfer'
selection_ps_platform:
CommandLine|contains:
- 'dropbox'
- 'gitlab'
- 'githubusercontent'
- 'github.com'
- 'pubnub'
- '104.253.175.177'
- '31.59.138.144'
condition: selection_mshta or selection_wscript or (selection_ps and selection_ps_download and selection_ps_platform)
falsepositives:
- Administrators running mshta/HTA-based internal tooling
- Legitimate logon or inventory scripts run by wscript from AppData or ProgramData
- Developers or admins using PowerShell to fetch scripts from GitHub or GitLab
level: high
Known false positives:
- Administrators running mshta/HTA-based internal tooling
- Legitimate logon scripts executed by wscript from ProgramData or AppData
- Developers or admins using PowerShell to pull scripts from GitHub or GitLab
Tuning notes: Scope to non-developer, non-admin hosts first. Allow-list known internal HTA/logon-script hashes and paths, and exclude approved GitHub/GitLab automation accounts. Adjust the index and sourcetype to your Sysmon ingestion; enable Sysmon EventCode 1 with command-line logging (or Windows 4688 with command-line auditing).
- HWP Lure Spawning [indicator] with Side-Loaded [indicator] — Exploitation Phase (Type G) · unlock with TL Creds
- caspol.exe Process Injection, Script-Host Launch and Network Activity — Post-Exploitation (Xctdoor) · unlock with TL Creds
- ASEC October 2026 Campaign IOC Sweep — C2/Staging URLs, IPs and MD5 Hashes — C2 Phase · unlock with TL Creds
- LNK-Launched PowerShell Decoding Embedded Payload (Hex/Base64/XOR) to VBS or Temp — Exploitation Phase · unlock with TL Creds
- AutoIt or Portable Python Interpreter Running from User-Writable Path — Post-Exploitation Backdoor Execution · unlock with TL Creds
- Unsanctioned MeshAgent Remote Management Tool Deployment — C2 / Remote Access · unlock with TL Creds
- Scheduled Task Masquerading as Chrome Update and Run-Key Script Launcher — Post-Exploitation Persistence · unlock with TL Creds
- Script Hosts and caspol.exe Reaching PubNub, Dropbox API, GitHub/GitLab or Campaign IPs — C2 and Exfiltration · unlock with TL Creds
Indicators of compromise
As of 2026-11-20, Threadlinqs Intelligence tracks 23 indicators of compromise for TL-2026-6549.
Sample indicators (3 of 23)
- IP address:
104.253.175[.]177 - File hash:
00fb5037e6a111c55f9d6673a103bf0a - IP address:
31.59.138[.]144
Defanged, TLP:CLEAR. Full set of 23 indicators: unlock with TL Creds.
Community OSINT corroboration
3 of this threat's indicators have also been reported by the open-source security community. Community sightings are unverified and are kept separate from Threadlinqs' curated indicators. Indicator values, reporters and campaign linkage are available to authenticated Red-tier users.
Technical analysis
The report body does not state attribution.
AhnLab Security Intelligence Center (ASEC) monitored APT attacks against Korean targets during October 2026 and published its monthly trend report on 2026-11-20. Most attacks were delivered by spear phishing with work-themed lures, and malicious LNK shortcut files had the largest share of all delivery types (the report gives no numeric statistics). The report groups the confirmed attacks into types by execution flow.
Type A (LNK): a PowerShell command inside the LNK extracts HEX data and restores a decoy document plus a malicious VBS file; persistence is a Run registry value; commands are received and results exfiltrated through a PubNub channel. Type B (LNK hosted on GitLab): a document-disguised LNK runs mshta, VBScript, PowerShell, Dropbox-hosted content or a DLL-based loader; the malware collects browser login data, cookies, cryptocurrency wallet data, keystrokes and clipboard contents, and abuses the legitimate remote-management tool MeshAgent. Type C (LNK, GitHub): LNK PowerShell writes Base64 data in %temp%, downloads a decoy and further scripts from GitHub, creates a scheduled task and deploys XenoRAT.
Type D (LNK, Dropbox): the LNK, VBS and Python files from Dropbox and an AES-encrypted DLL are used to inject code into caspol.exe, after which Xctdoor is loaded in memory and performs command execution and screen capture over HTTP. Type E (PowerShell): Base64 PowerShell decodes a decoy and malicious scripts with XOR, registers a scheduled task disguised as a Google Chrome update and loads a .NET-based RAT in memory for remote control. Type F (LNK, Dropbox API): the LNK restores a decoy and an XOR-encoded ZIP containing a legitimate AutoIt interpreter and a malicious script; system information is exfiltrated through the Dropbox API and a BAT command file is then downloaded and run (backdoor function). Type G (HWP): a Hangul document disguised as defense-industry work material; double-clicking the attached icon runs WinObj, which side-loads a malicious version.dll; a scheduled task then runs a PowerShell backdoor that sends system information and the process list to a C2 server and supports remote command execution and file transfer. An additional 'Unknown' type was delivered by spear phishing but not detailed.
Attribution: the report body contains no attribution statement. The page's tag list includes 'Kimsuky' and 'kimjongrat', but these are tags only and are not treated here as confirmed attribution. Separately, earlier ASEC research (Aug 2026) attributes Xctdoor distribution to the Larva-26005 cluster, which ASEC links to CRAT; that is background and not a stated link for the October activity. No CVE or vulnerability exploitation is described; the page publishes five MD5 hashes and five URLs on 104.253.175.177 and 31.59.138.144. BeaconBeagle returned HTTP 404 (no record) for both IPs at research time. Severity is an analyst judgement.
MITRE ATT&CK techniques by tactic:
Defense Evasion
T1036.004 Masquerade Task or Service; T1055 Process Injection; T1140 Deobfuscate/Decode Files or Information; T1218.005 Mshta
Persistence
T1053.005 Scheduled Task; T1547.001 Registry Run Keys / Startup Folder
Collection
T1056.001 Keylogging; T1113 Screen Capture
Discovery
T1057 Process Discovery; T1082 System Information Discovery
Execution
T1059.001 PowerShell; T1059.005 Visual Basic; T1059.006 Python; T1204.002 Malicious File
Command and Control
Credential Access
T1539 Steal Web Session Cookie; T1555.003 Credentials from Web Browsers
Initial Access
T1566.001 Spearphishing Attachment
Exfiltration
T1567.002 Exfiltration to Cloud Storage
stealth
Remediation
Patches
- No CVE or vendor patch is associated; keep OS, browsers and endpoint security (e.g. V3) updated
Immediate actions
- Block outbound access to 104.253.175.177 and 31.59.138.144 and the listed URLs; hunt proxy/DNS logs for them
- Search endpoints for the five listed MD5 hashes
- Hunt for LNK files launching powershell.exe, mshta.exe, wscript.exe or python from user-writable paths, and for caspol.exe with network activity or injected modules
- Hunt for WinObj executed from non-standard directories alongside a version.dll, and for scheduled tasks mimicking Google Chrome updates
- Review endpoints for unsanctioned MeshAgent installs
Workarounds
- Block or open-in-sandbox .lnk attachments at the mail gateway
- Disable mshta.exe where not required
- Show file extensions to expose double-extension shortcut lures such as .hwp.lnk
Longer-term hardening
- Deploy EDR with behavioral detection for script-host and LNK abuse
- Alert on anomalous access to PubNub, Dropbox API, GitHub raw/GitLab content from non-developer hosts
- Train users on work-themed spear phishing with shortcut and Hangul (HWP) attachments
- Restrict script interpreters (mshta, wscript, AutoIt, Python) via application control
Affected and fixed versions
- Microsoft — Windows (endpoints receiving LNK/HWP spear-phishing lures)
- Hancom — Hangul (HWP) documents used as lure format; no vulnerability cited
Timeline
- Hauri reported an Xctdoor attack using a Veraport-disguised installer (month-level date per ASEC Aug 2026 analysis; exact day not stated)
- ASEC observed LNK-based Xctdoor attacks starting in April 2026, recurring in June and July 2026 (month-level date; exact day not stated)
- ASEC published the May 2026 APT trend report (Korea) covering LNK/JSE/CHM lures, XenoRAT-type malware, Python and AutoIt backdoors and MeshAgent-related detections
- ASEC published its analysis of Xctdoor and its connection to past CRAT cases (Larva-26005); date as shown on the ASEC English page
- ASEC published the July 2026 APT trend report (Korea) describing LNK-based spear-phishing Types A-G using AutoIt, PowerShell, HTA, Python backdoors and DLL side-loading
- Start of the October 2026 window in which ASEC monitored APT attacks against Korean targets
- End of the October 2026 monitoring window; LNK files were the largest share of spear-phishing delivery
- ASEC published the October 2026 APT attack trend report (Korea) with Types A-G, five MD5 hashes and five C2/staging URLs
Sources cited
- 2026년 10월 APT 공격 동향 보고서(국내) - AhnLab ASEC asec.ahnlab.com · Research · 2026-11-20
- Analysis of the Connection Between Xctdoor and Past CRAT Attack Cases (Larva-26005) - AhnLab ASEC asec.ahnlab.com · Research · 2026-08-04
- July 2026 Threat Trend Report on APT Attacks (South Korea) - AhnLab ASEC asec.ahnlab.com · Research · 2026-08-28
- May 2026 Threat Trend Report on APT Attacks (South Korea) - AhnLab ASEC asec.ahnlab.com · Research · 2026-06-25
- Larva-26005 actor profile - Malpedia malpedia.caad.fkie.fraunhofer.de · Research
- Larva-26005: Tracing Xctdoor's Connection to CRAT Attacks - SOC Prime socprime.com · Blog
- South Korea APT Threat Report August 2026 Attack Trends - SOC Prime socprime.com · Blog
Drafted by the Threadlinqs research pipeline from 7 sources; published ; last updated —.