Skip to content

Command

Summary

Matches if a command finishes with a successful exit code and the configured output. Command conditions also run with --dry-run.

Example

rules:
  # Match if the script exits with 0 or 2 and prints "successful" (case-sensitive)
  - name: "Check is successful"
    conditions:
      command:
        run: '.\my-check.ps1'
        working_directory: scripts # Relative to the policy file
        success_exit_codes: [0, 2]
        stdout:
          value: "successful"
          match:
            type: contains
        trusted_files:
          - scripts/my-check.ps1 # Relative to the policy file (not to working_directory)
    actions:
      - run: "echo Check is successful"
rules:
  # Match if the script exits with 0 or 2 and prints "successful" (case-insensitive)
  - name: "Check is successful"
    conditions:
      command:
        run: "./my-check.sh"
        working_directory: scripts # Relative to the policy file
        success_exit_codes: [0, 2]
        stdout:
          value: "successful"
          match:
            type: contains
            case_sensitive: false
        trusted_files:
          - scripts/my-check.sh # Relative to the policy file (not to working_directory)
    actions:
      - run: "echo Check is successful"

Options

Command conditions use the command options.

Details

Warning

Command conditions run on every evaluation, even with --dry-run. Use trust to detect modifications.

  • The condition matches if the exit code and all configured outputs match. A different exit code or output is no error, the condition simply does not match.
  • If the command cannot be started or exceeds its timeout, the condition cannot be evaluated and the evaluation stops.
  • Unlike actions, conditions do not receive the POWERRULES_* environment variables.
  • Commands and their output are only logged on DEBUG level.
  • and and or stop at the first decisive result. Put cheap conditions first, so the command only runs when needed.

See Actions vs. command conditions for the differences to the commands of actions.

  • Policy Evaluation


    Continuous evaluation and --dry-run.

  • Actions


    Run commands when a rule matches.

  • Trust


    Detect modifications of policies and scripts.