Interactive Console
The Interactive Console runs commands on one machine through its UTMStack agent, without opening SSH or Remote Desktop. Use it to investigate a machine, contain a threat by hand, or test a command before you put it in a flow.
Open the console
You can open it in two places:
SOAR > Interactive Console. The left list shows your agents with a search box (Search agents…). Select an agent to open its console. Until you do, the page reads "Select an agent — Choose an agent from the list to open its interactive console."
From a host in Configure > Data Sources: the Console section ("Run commands on this host through its agent (live response).") has an Open console button.
Who can use it: users with the soar.write permission (Administrator and Analyst roles). Viewers are refused.
Run commands
- 1
Select the agent
Agents installed with remote control disabled show a lock and "Remote control disabled". They refuse every command, and this can only be changed by reinstalling the agent on that machine.
- 2
Confirm your password
Before the first command, the console asks you to authenticate again ("Authenticate to run commands on this agent."). Type your UTMStack password. The characters are shown as dots. A wrong password shows "invalid credentials"; after too many failures the console shows "too many failed attempts; try again later".
Only accounts with a UTMStack password can pass this step. Accounts that sign in through single sign-on or a directory (LDAP) have no UTMStack password and cannot use the console. Failed attempts count toward the sign-in lockout: after 9 failures within 10 minutes from one address, sign-in from that address is blocked for 10 minutes.
- 3
Choose the shell
On Windows, use Shell type to pick
cmd(the default) orpowershell. On Linux and macOS the console always usesbash. - 4
Type a command and press Enter
"Type a command and press Enter to run it on the agent." The output appears below the command when it finishes.
If the agent is offline, the console shows "This agent appears offline — it may not respond to commands." and the command box reads "Agent offline — commands are disabled". For an agent installed with remote control disabled, the console explains: "This agent was installed with remote control disabled. It refuses commands, and that can only be changed on the machine itself."
The password proof is valid for 5 minutes. If a command is refused after the console has been open for a while, close the console and open it again to authenticate again.
Use variables
The Variables button opens Command variables, where you create, edit and delete stored variables and insert them at the cursor. A variable is written $[variables.NAME] and is filled in when the command runs. Secret values are replaced by **** in the output. See Step Outputs and Variables.
How commands run
Each command runs once and returns its full output when it finishes. Output is not streamed line by line.
The command runs with the agent's own rights: root on Linux, the Local System account on Windows. The shell is started the same way as for flow steps (
cmd.exe /C,powershell.exe -Command,bash -corsh -c; see Step Types).Commands for one agent run one at a time. A long console command delays flow steps on the same machine, and the other way round.
Agents from UTMStack v12.0.2 cut output longer than 65,536 bytes and add the line
... output truncated, showing first N of M bytes. Older agents return the error "string exceeds 65536 characters" instead.When a command fails, the output holds what it printed, its error text, and a last line such as
exit status 1.On Windows, text in other alphabets may appear garbled in the output.
Interactive programs that wait for input (for example
passwdwithout options, or a command that asks "Are you sure?") never get an answer. Use options that make the command run without questions.
Where commands are recorded
Every console command is stored in SOAR > Execution history with the origin Manual and the agent it ran on. The Node column shows "Interactive console" and the Flow column shows who ran the command. A console command that ended with an error is still recorded as Executed; only a broken connection to the agent records Failed. See Execution History and Run Statuses.
Use Execution history as the record of console commands. In testing of UTMStack v12.0.1, console commands and variable changes did not appear in Configure > Settings > App Audit Logs.
Use cases
Investigation: list processes, connections or users on a machine that raised an alert.
Manual containment: block an address or disable an account right away when no flow covers the alert.
Testing: run a command by hand, with the same shell, before you put it in a flow.
Troubleshooting: check that an agent is online and accepts commands.
Changed from UTMStack v11
The console now asks for your password before you run commands.
Automation Variables is no longer a separate page; use the Variables button.
SOAR > Audit is gone; the old address opens Configure > Settings > App Audit Logs. Console commands are listed in Execution history.
v11 refused commands that printed a secret. UTMStack v12 runs them and masks the secret in the output.
Related pages
Step Outputs and Variables
Execution History and Run Statuses
Running Steps on Another Host
Troubleshooting Flows