Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Adversaries may transfer tools or other files between systems in a compromised environment.
.claude/skills/cyberstrikeus-t1570-lateral-tool-transfer/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 4% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 69% | 0% |
Adversaries may transfer tools or other files between systems in a compromised environment. Once brought into the victim environment (i.e., Ingress Tool Transfer) files may then be copied from one system to another to stage adversary tools or other files over the course of an operation.
Adversaries may copy files between internal victim systems to support lateral movement using inherent file sharing protocols such as file sharing over SMB/Windows Admin Shares to connected network shares or with authenticated connections via Remote Desktop Protocol.
Files can also be transferred using native or otherwise present tools on the victim system, such as scp, rsync, curl, sftp, and ftp. In some cases, adversaries may be able to leverage Web Services such as Dropbox or OneDrive to copy files from one machine to another via shared, automatically synced folders.
Platforms: ESXi, Linux, macOS, Windows
The following tests are from Atomic Red Team and provide actionable ways to test this technique:
Simulates an attacker exfiltrating data over SMB over QUIC using the New-SmbMapping command. Prerequisites:
Supported Platforms: windows Elevation Required: Yes
powershellNew-SmbMapping -RemotePath '#{remote_path}' -TransportType QUIC -SkipCertificateCheck copy '#{local_file}' 'Z:\'
Simulates an attacker exfiltrating data over SMB over QUIC using the NET USE command. Prerequisites:
Supported Platforms: windows Elevation Required: Yes
powershellNET USE * '#{remote_path}' /TRANSPORT:QUIC /SKIPCERTCHECK copy '#{local_file}' '*:\'
If Atomic Red Team tests are not applicable, manually verify the technique by:
Consider using the host firewall to restrict file sharing communications such as SMB.
Network intrusion detection and prevention systems that use network signatures to identify traffic for specific adversary malware or unusual data transfer over known tools and protocols like FTP 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.
| Finding | Severity | Impact | | ------------------------------------------ | -------- | ---------------- | | Lateral Tool Transfer technique applicable | Low | Lateral Movement |
| CWE ID | Title | | ------- | ----------------------- | | CWE-284 | Improper Access Control |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 26,514 | 22,622 | -15% | 1 | 1 | 0% | 3,798 | 4,271 | +12% | 0 | 0 | — |
case-02 | fail→pass | 37,588 | 26,171 | -30% | 1 | 1 | 0% | 5,554 | 5,784 | +4% | 0 | 0 | — |
case-03 | fail→fail | 22,671 | 28,793 | +27% | 1 | 1 | 0% | 3,002 | 4,532 | +51% | 0 | 0 | — |
case-04 | fail→pass | 18,511 | 15,036 | -19% | 1 | 1 | 0% | 2,029 | 2,476 | +22% | 0 | 0 | — |
case-05 | pass→pass | 13,466 | 11,819 | -12% | 1 | 1 | 0% | 1,470 | 1,922 | +31% | 0 | 0 | — |
case-06 | pass→pass | 25,642 | 15,977 | -38% | 1 | 1 | 0% | 2,864 | 3,012 | +5% | 0 | 0 | — |
case-07 | fail→pass | 15,912 | 11,516 | -28% | 1 | 1 | 0% | 1,877 | 1,855 | -1% | 0 | 0 | — |
case-08 | pass→pass | 17,251 | 12,785 | -26% | 1 | 1 | 0% | 1,781 | 2,439 | +37% | 0 | 0 | — |
case-09 | pass→pass | 15,514 | 8,403 | -46% | 1 | 1 | 0% | 1,835 | 1,291 | -30% | 0 | 0 | — |
case-10 | fail→pass | 9,008 | 9,583 | +6% | 1 | 1 | 0% | 1,345 | 1,393 | +4% | 0 | 0 | — |
case-11 | pass→pass | 12,398 | 49,315 | +298% | 1 | 1 | 0% | 1,289 | 1,569 | +22% | 0 | 0 | — |
case-12 | pass→pass | 17,872 | 9,913 | -45% | 1 | 1 | 0% | 2,033 | 1,708 | -16% | 0 | 0 | — |
case-13 | fail→pass | 17,175 | 19,183 | +12% | 1 | 1 | 0% | 2,332 | 3,942 | +69% | 0 | 0 | — |
case-14 | pass→pass | 23,728 | 25,944 | +9% | 1 | 1 | 0% | 2,913 | 5,326 | +83% | 0 | 0 | — |
case-15 | fail→pass | 19,517 | 8,929 | -54% | 1 | 1 | 0% | 2,114 | 1,454 | -31% | 0 | 0 | — |
case-16 | fail→pass | 22,849 | 25,949 | +14% | 1 | 1 | 0% | 3,250 | 4,019 | +24% | 0 | 0 | — |
case-17 | pass→pass | 11,629 | 7,265 | -38% | 1 | 1 | 0% | 2,161 | 1,474 | -32% | 0 | 0 | — |
case-18 | pass→pass | 7,860 | 10,532 | +34% | 1 | 1 | 0% | 1,378 | 1,562 | +13% | 0 | 0 | — |
case-19 | pass→pass | 21,212 | 2,740 | -87% | 1 | 1 | 0% | 809 | 1,445 | +79% | 0 | 0 | — |
case-20 | pass→pass | 13,865 | 14,436 | +4% | 1 | 1 | 0% | 1,666 | 2,609 | +57% | 0 | 0 | — |
case-21 | pass→pass | 24,245 | 19,199 | -21% | 1 | 1 | 0% | 2,630 | 4,625 | +76% | 0 | 0 | — |
case-22 | pass→pass | 20,483 | 21,167 | +3% | 1 | 1 | 0% | 2,771 | 3,419 | +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. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 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.