skills-cerewro
Español

Command Executor Skill: security and control on every execution

Cerewro's Command Executor Skill runs commands with minimum required privilege, shows intent before launching and requires approval for critical actions. Prevents destructive errors.

Equipo Cerewro
skill run_command seguridad aprobaciones

Command Executor Skill: security first

Cerewro's Command Executor Skill defines how the AI behaves when running system commands. Its core principle is least privilege: it only requests elevation when absolutely necessary.

Skill behavior

  1. Shows intent: before executing, describes what it's going to do and why
  2. Requests confirmation: for critical operations (deletion, service stopping), requires approval
  3. Least privilege: avoids elevating privileges unless strictly necessary
  4. Reports result: shows command output in a readable format
Example: confirmation before executing
I'm about to run: Stop-Service -Name "W3SVC"
This will stop the IIS service. Do you confirm? (yes/no)
Why it matters: In production environments, a badly executed command can take down a critical service. The Command Executor Skill acts as a human verification layer before any action with real impact.