Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Adversaries may stop or disable services on a system to render those services unavailable to legitimate users.
.claude/skills/cyberstrikeus-t1489-service-stop/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 64% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 25% | 0% |
| case-22 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 42% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 63% | 0% |
Adversaries may stop or disable services on a system to render those services unavailable to legitimate users. Stopping critical services or processes can inhibit or stop response to an incident or aid in the adversary's overall objectives to cause damage to the environment.
Adversaries may accomplish this by disabling individual services of high importance to an organization, such as <code>MSExchangeIS</code>, which will make Exchange content inaccessible. In some cases, adversaries may stop or disable many or all services to render systems unusable. Services or processes may not allow for modification of their data stores while running. Adversaries may stop services or processes in order to conduct Data Destruction or Data Encrypted for Impact on the data stores of services like Exchange and SQL Server, or on virtual machines hosted on ESXi infrastructure.
Threat actors may also disable or stop service in cloud environments. For example, by leveraging the DisableAPIServiceAccess API in AWS, a threat actor may prevent the service from creating service-linked roles on new accounts in the AWS Organization.
Platforms: ESXi, IaaS, Linux, macOS, Windows
The following tests are from Atomic Red Team and provide actionable ways to test this technique:
Stops a specified service using the sc.exe command. Upon execution, if the spooler service was running infomration will be displayed saying it has changed to a state of STOP_PENDING. If the spooler service was not running "The service has not been started." will be displayed and it can be started by running the cleanup command.
Supported Platforms: windows Elevation Required: Yes
cmdsc.exe stop #{service_name}
Stops a specified service using the net.exe command. Upon execution, if the service was running "The Print Spooler service was stopped successfully." will be displayed. If the service was not running, "The Print Spooler service is not started." will be displayed and it can be started by running the cleanup command.
Supported Platforms: windows Elevation Required: Yes
cmdnet.exe stop #{service_name}
Stops a specified service killng the service's process. This technique was used by WannaCry. Upon execution, if the spoolsv service was running "SUCCESS: The process "spoolsv.exe" with PID 2316 has been terminated." will be displayed. If the service was not running "ERROR: The process "spoolsv.exe" not found." will be displayed and it can be started by running the cleanup command.
Supported Platforms: windows
cmdtaskkill.exe /f /im #{process_name}
Stops a specified service using the systemctl command. Upon execution, if the specified service was running, it will change to a state of inactive and it can be restarted by running the cleanup command. You can list all available services with following command: "systemctl list-units --type=service"
Supported Platforms: linux Elevation Required: Yes
bashsudo systemctl stop #{service_name}
Stops a specified service by sending a SIGTERM signal to the linked process using the killall command. Upon execution, if the service's main process was running, it will be terminated. If the service was not running, no process will be found to kill and it can be restarted by running the cleanup command. You can list all available services with following command: "systemctl list-units --type=service"
Supported Platforms: linux Elevation Required: Yes
bashsudo killall -SIGTERM #{process_name}
If Atomic Red Team tests are not applicable, manually verify the technique by:
Operate intrusion detection, analysis, and response systems on a separate network from the production environment to lessen the chances that an adversary can see and interfere with critical response functions.
Limit privileges of user accounts and groups so that only authorized administrators can interact with service changes and service configurations.
Develop and enforce security policies that include the use of out-of-band communication channels for critical communications during a security incident.
Ensure proper registry permissions are in place to inhibit adversaries from disabling or interfering with critical services.
Ensure proper process and file permissions are in place to inhibit adversaries from disabling or interfering with critical services.
| Finding | Severity | Impact | | --------------------------------- | -------- | ------ | | Service Stop 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-01 | fail→pass | 20,194 | 20,493 | +1% | 1 | 1 | 0% | 2,636 | 4,311 | +64% | 0 | 0 | — |
case-02 | fail→fail | 18,258 | 20,693 | +13% | 1 | 1 | 0% | 3,163 | 4,503 | +42% | 0 | 0 | — |
case-03 | fail→fail | 37,044 | 31,804 | -14% | 1 | 1 | 0% | 5,635 | 6,753 | +20% | 0 | 0 | — |
case-04 | pass→pass | 19,000 | 11,202 | -41% | 1 | 1 | 0% | 2,554 | 3,631 | +42% | 0 | 0 | — |
case-05 | pass→pass | 14,347 | 19,051 | +33% | 1 | 1 | 0% | 2,712 | 4,422 | +63% | 0 | 0 | — |
case-06 | pass→pass | 27,498 | 24,429 | -11% | 1 | 1 | 0% | 4,092 | 5,124 | +25% | 0 | 0 | — |
case-07 | pass→pass | 12,181 | 10,391 | -15% | 1 | 1 | 0% | 818 | 2,104 | +157% | 0 | 0 | — |
case-08 | pass→pass | 16,230 | 10,455 | -36% | 1 | 1 | 0% | 1,737 | 2,394 | +38% | 0 | 0 | — |
case-09 | pass→pass | 11,230 | 9,009 | -20% | 1 | 1 | 0% | 1,288 | 2,365 | +84% | 0 | 0 | — |
case-10 | pass→pass | 5,578 | 2,628 | -53% | 1 | 1 | 0% | 785 | 2,106 | +168% | 0 | 0 | — |
case-11 | pass→pass | 10,042 | 9,192 | -8% | 1 | 1 | 0% | 805 | 2,150 | +167% | 0 | 0 | — |
case-12 | pass→pass | 12,131 | 12,869 | +6% | 1 | 1 | 0% | 965 | 2,447 | +154% | 0 | 0 | — |
case-13 | pass→pass | 9,897 | 7,180 | -27% | 1 | 1 | 0% | 1,482 | 1,985 | +34% | 0 | 0 | — |
case-14 | pass→pass | 20,188 | 26,447 | +31% | 1 | 1 | 0% | 2,874 | 4,555 | +58% | 0 | 0 | — |
case-15 | pass→pass | 24,759 | 18,776 | -24% | 1 | 1 | 0% | 3,038 | 4,383 | +44% | 0 | 0 | — |
case-16 | pass→pass | 15,677 | 10,367 | -34% | 1 | 1 | 0% | 1,670 | 2,272 | +36% | 0 | 0 | — |
case-17 | pass→pass | 12,641 | 8,024 | -37% | 1 | 1 | 0% | 1,170 | 2,213 | +89% | 0 | 0 | — |
case-18 | pass→pass | 11,958 | 8,417 | -30% | 1 | 1 | 0% | 1,003 | 2,372 | +136% | 0 | 0 | — |
case-19 | pass→pass | 9,229 | 7,232 | -22% | 1 | 1 | 0% | 715 | 2,018 | +182% | 0 | 0 | — |
case-20 | pass→pass | 21,936 | 13,412 | -39% | 1 | 1 | 0% | 2,863 | 3,188 | +11% | 0 | 0 | — |
case-21 | fail→pass | 17,098 | 7,252 | -58% | 1 | 1 | 0% | 1,761 | 2,208 | +25% | 0 | 0 | — |
case-22 | fail→pass | 17,496 | 7,516 | -57% | 1 | 1 | 0% | 2,100 | 2,169 | +3% | 0 | 0 | — |
case-23 | pass→pass | 16,178 | 21,664 | +34% | 1 | 1 | 0% | 2,373 | 3,986 | +68% | 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. 23 cases were attempted. The headline lift of +13 percentage points is the difference between those two pass rates over the 23 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.