Install any skill in seconds. Free to start, no credit card required.
Get Started Free →An adversary may deface systems internal to an organization in an attempt to intimidate or mislead users, thus discrediting the integrity of the systems.
.claude/skills/cyberstrikeus-t1491-001-internal-defacement/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-11 | ✗→✓ | ▲ Improved | 29% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 139% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 113% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 62% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 163% | 0% |
> Sub-technique of: T1491
An adversary may deface systems internal to an organization in an attempt to intimidate or mislead users, thus discrediting the integrity of the systems. This may take the form of modifications to internal websites or server login messages, or directly to user systems with the replacement of the desktop wallpaper. Disturbing or offensive images may be used as a part of Internal Defacement in order to cause user discomfort, or to pressure compliance with accompanying messages. Since internally defacing systems exposes an adversary's presence, it often takes place after other intrusion goals have been accomplished.
Platforms: ESXi, Linux, macOS, Windows
The following tests are from Atomic Red Team and provide actionable ways to test this technique:
Downloads an image from a URL and sets it as the desktop wallpaper.
Supported Platforms: windows
powershell$url = "#{url_of_wallpaper}" $imgLocation = "#{wallpaper_location}" $orgWallpaper = (Get-ItemProperty -Path Registry::'HKEY_CURRENT_USER\Control Panel\Desktop\' -Name WallPaper).WallPaper $orgWallpaper | Out-File -FilePath "#{pointer_to_orginal_wallpaper}" $updateWallpapercode = @' using System.Runtime.InteropServices; namespace Win32{ public class Wallpaper{ [DllImport("user32.dll", CharSet=CharSet.Auto)] static extern int SystemParametersInfo (int uAction , int uParam , string lpvParam , int fuWinIni) ; public static void SetWallpaper(string thePath){ SystemParametersInfo(20,0,thePath,3); } } } '@ $wc = New-Object System.Net.WebClient try{ $wc.DownloadFile($url, $imgLocation) add-type $updateWallpapercode [Win32.Wallpaper]::SetWallpaper($imgLocation) } catch [System.Net.WebException]{ Write-Host("Cannot download $url") add-type $updateWallpapercode [Win32.Wallpaper]::SetWallpaper($imgLocation) } finally{ $wc.Dispose() }
Display ransom message to users at system start-up by configuring registry keys HKLM\SOFTWARE\Micosoft\Windows\CurrentVersion\Policies\System\LegalNoticeCaption and HKLM\SOFTWARE\Micosoft\Windows\CurrentVersion\Policies\System\LegalNoticeText.
SynAck Ransomware, Grief Ransomware, Maze Ransomware, Pysa Ransomware, Spook Ransomware, DopplePaymer Ransomware, Reedemer Ransomware, Kangaroo Ransomware
Supported Platforms: windows Elevation Required: Yes
powershell$orgLegalNoticeCaption = (Get-ItemProperty HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System -Name LegalNoticeCaption).LegalNoticeCaption $orgLegalNoticeText = (Get-ItemProperty HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System -Name LegalNoticeText).LegalNoticeText $newLegalNoticeCaption = "#{legal_notice_caption}" $newLegalNoticeText = "#{legal_notice_text}" Set-ItemProperty HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System -Name LegalNoticeCaption -Value $newLegalNoticeCaption -Type String -Force Set-ItemProperty HKLM:\SOFTWARE\Microsoft\Windows\CurrentVersion\Policies\System -Name LegalNoticeText -Value $newLegalNoticeText -Type String -Force
Changes the ESXi welcome message to potentially display ransom information. Reference
Supported Platforms: windows
cmdecho "" | "#{plink_file}" -batch "#{vm_host}" -ssh -l #{vm_user} -pw "#{vm_pass}" "esxcli system welcomemsg set -m 'RANSOMWARE-NOTIFICATION'"
Dependencies:
Creates a temporary ransom-note text file and opens it in Notepad to simulate ransomware "note display" behavior without making destructive changes. SAFE and non-destructive.
Supported Platforms: windows
powershell$notePath = Join-Path $env:TEMP "#{note_filename}" $pidPath = Join-Path $env:TEMP "#{pid_filename}" $Title = "#{note_title}" $Body = "#{note_body}" $header = $Title + "`r`n" + ('=' * $Title.Length) + "`r`n`r`n" $content = $header + $Body [System.IO.File]::WriteAllText($notePath, $content, [System.Text.Encoding]::UTF8) $p = Start-Process notepad.exe -ArgumentList "`"$notePath`"" -PassThru $p.Id | Out-File -FilePath $pidPath -Encoding ascii -Force
Dependencies:
If Atomic Red Team tests are not applicable, manually verify the technique by:
Consider implementing IT disaster recovery plans that contain procedures for taking regular data backups that can be used to restore organizational data. Ensure backups are stored off system and is protected from common methods adversaries may use to gain access and destroy the backups to prevent recovery.
| Finding | Severity | Impact | | ---------------------------------------- | -------- | ------ | | Internal Defacement technique applicable | Low | Impact |
| CWE ID | Title | | ------- | --------------------------------- | | CWE-400 | Uncontrolled Resource Consumption |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | fail→pass | 11,227 | 6,922 | -38% | 1 | 1 | 0% | 2,013 | 2,601 | +29% | 0 | 0 | — |
case-01 | fail→pass | 14,565 | 13,573 | -7% | 1 | 1 | 0% | 1,573 | 3,756 | +139% | 0 | 0 | — |
case-02 | fail→pass | 16,563 | 12,234 | -26% | 1 | 1 | 0% | 2,117 | 4,518 | +113% | 0 | 0 | — |
case-17 | fail→pass | 21,216 | 17,037 | -20% | 1 | 1 | 0% | 2,443 | 3,966 | +62% | 0 | 0 | — |
case-03 | pass→pass | 6,341 | 9,061 | +43% | 1 | 1 | 0% | 1,055 | 3,078 | +192% | 0 | 0 | — |
case-04 | pass→pass | 8,501 | 6,416 | -25% | 1 | 1 | 0% | 1,682 | 3,554 | +111% | 0 | 0 | — |
case-05 | pass→pass | 12,323 | 9,861 | -20% | 1 | 1 | 0% | 1,263 | 2,946 | +133% | 0 | 0 | — |
case-06 | fail→pass | 10,245 | 7,235 | -29% | 1 | 1 | 0% | 1,036 | 2,727 | +163% | 0 | 0 | — |
case-07 | pass→pass | 9,433 | 3,441 | -64% | 1 | 1 | 0% | 1,369 | 2,865 | +109% | 0 | 0 | — |
case-08 | fail→pass | 21,038 | 7,276 | -65% | 1 | 1 | 0% | 3,012 | 2,550 | -15% | 0 | 0 | — |
case-09 | fail→pass | 7,794 | 9,933 | +27% | 1 | 1 | 0% | 1,466 | 2,627 | +79% | 0 | 0 | — |
case-10 | pass→pass | 10,646 | 2,876 | -73% | 1 | 1 | 0% | 985 | 2,755 | +180% | 0 | 0 | — |
case-12 | fail→pass | 10,072 | 9,858 | -2% | 1 | 1 | 0% | 1,588 | 3,118 | +96% | 0 | 0 | — |
case-13 | pass→pass | 19,119 | 13,702 | -28% | 1 | 1 | 0% | 2,898 | 4,028 | +39% | 0 | 0 | — |
case-14 | fail→pass | 26,126 | 10,624 | -59% | 1 | 1 | 0% | 3,505 | 2,968 | -15% | 0 | 0 | — |
case-15 | fail→pass | 28,605 | 12,977 | -55% | 1 | 1 | 0% | 3,393 | 3,629 | +7% | 0 | 0 | — |
case-16 | pass→pass | 11,496 | 7,239 | -37% | 1 | 1 | 0% | 1,087 | 2,614 | +140% | 0 | 0 | — |
case-18 | pass→pass | 10,372 | 11,673 | +13% | 1 | 1 | 0% | 1,030 | 3,286 | +219% | 0 | 0 | — |
case-19 | pass→pass | 9,922 | 10,017 | +1% | 1 | 1 | 0% | 956 | 2,934 | +207% | 0 | 0 | — |
case-20 | pass→pass | 31,308 | 8,430 | -73% | 1 | 1 | 0% | 1,559 | 2,861 | +84% | 0 | 0 | — |
case-21 | pass→pass | 3,956 | 2,954 | -25% | 1 | 1 | 0% | 576 | 2,760 | +379% | 0 | 0 | — |
case-22 | pass→pass | 13,217 | 11,493 | -13% | 1 | 1 | 0% | 1,509 | 3,413 | +126% | 0 | 0 | — |
case-23 | pass→pass | 16,156 | 9,115 | -44% | 1 | 1 | 0% | 1,916 | 3,816 | +99% | 0 | 0 | — |
case-24 | pass→pass | 23,121 | 21,685 | -6% | 1 | 1 | 0% | 2,834 | 5,044 | +78% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted. The headline lift of +42 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.