SSH Command Node

The SSH Command node allows you to execute commands on remote servers via the SSH protocol. It supports multiple steps (sequential commands), authentication by password or private key, and output capture.

Overview

Property	Value
Type	`ssh-command`
Category	Infrastructure
Color	Light Blue (#0ea5e9)
Input	`in`
Output	`out`

Connection

Using Saved Credentials (Recommended)

Select a credential of type SSH in the Credential field.

Manual Configuration

Field	Type	Description
Host	`string`	Server address
Port	`number`	SSH port (default: 22)
Username	`string`	Username
Password	`string`	Password (password authentication)
Private Key	`string`	SSH private key (key-based authentication)
Passphrase	`string`	Key passphrase (if key is protected)

Authentication Methods

Method	Required Fields
Password	Username + Password
Private Key	Username + Private Key + Passphrase (optional)

Configuration

Global Timeout

Field	Type	Default	Description
Timeout (ms)	`number`	30000	Maximum time for total execution

Steps (SSH Steps)

The node supports multiple steps executed sequentially within the same SSH connection:

Field	Type	Description
Label	`string`	Descriptive name for the step
Command	`string`	Command to execute (supports `{{ }}`)
Wait for Match	`boolean`	Wait for text to appear in the output
Match String	`string`	Text to wait for
Use Regex	`boolean`	Interpret match as a regular expression
Match Timeout (ms)	`number`	Maximum time to wait for match

Wait for Match

The Wait for Match feature allows a step to wait until specific text appears in the command output. Useful for:

Waiting for services to start
Waiting for interactive prompts
Confirming that an operation has completed

Example:

Command: systemctl restart nginx
Wait for Match: true
Match String: "Active: active (running)"
Timeout: 10000

Outputs

Output	Type	Description
`stdout`	`string`	Complete standard output
`stderr`	`string`	Error output
`exitCode`	`number`	Exit code (0 = success)
`steps`	`array`	Result of each individual step

Accessing Outputs

// Full output
{{ steps.ssh.outputs.stdout }}

// Exit code
{{ steps.ssh.outputs.exitCode }}

// Result of a specific step
{{ steps.ssh.outputs.steps[0].stdout }}
{{ steps.ssh.outputs.steps[0].exitCode }}

Practical Examples

Check service status

Step 1:
  Label: Check Nginx
  Command: systemctl status nginx

Application deployment

Step 1:
  Label: Go to directory
  Command: cd /opt/minha-app

Step 2:
  Label: Pull repository
  Command: git pull origin main

Step 3:
  Label: Install dependencies
  Command: npm install --production

Step 4:
  Label: Restart service
  Command: pm2 restart minha-app
  Wait for Match: true
  Match String: "online"

Collect server information

Step 1:
  Label: Disk usage
  Command: df -h /

Step 2:
  Label: Memory
  Command: free -m

Step 3:
  Label: Processes
  Command: ps aux --sort=-%mem | head -10

Flow with Validation

[SSH Command: "deploy"]
    │
    ▼
[If: {{ steps.ssh.outputs.exitCode }} === 0]
    │ true → [HTTP Request: GET /health-check]
    │            │
    │            ▼
    │         [Assert: status === 200]
    │
    │ false → [Log: "Deploy falhou: {{ steps.ssh.outputs.stderr }}"]
              → [Stop and Fail]

Tips

Use saved credentials to avoid exposing passwords/keys in flows
Check the exitCode — 0 indicates success, other values indicate an error
Use expressions in commands: echo "Deploy em {{ variables.ENVIRONMENT }}"
Wait for Match is essential for long-running commands (restarts, builds)
All steps run within the same SSH session — the working directory is preserved between steps
Commands with {{ }} are evaluated before execution