SSH Client
SSH Client connects to one of your servers over SSH and runs the command you give it, such as checking free space or a service’s status. It returns what the server answers, formatted as a table when appropriate.
When to use it
- Checking a server’s free space every morning and alerting if it is running low.
- Letting a tech support agent check a service’s status.
- Reviewing the last lines of an error log.
Configuration
You need a SSH Client credential with the server address, port, user and password or key. See Credentials.
| Field | What it does |
|---|---|
| Credential (required) | The SSH Client credential. |
| Operation (required) | Execute Command, the only one available. |
| Command (required) | The command to run, for example df -h. |
| Output Format | How the answer is presented: Auto (default) turns it into a table if it is one; Plain Text keeps it as text with its line breaks; Markdown turns it into a table. |
Advanced options
| Field | What it does |
|---|---|
| Timeout (Seconds) | The longest the command is waited for before it is stopped. Default: 10. Raise it for slow commands. |
Output
The Message output is the server’s answer.
In tool mode, an agent can use the component by itself. See Tool Mode.
Example
A systems team wants a daily disk space alert.
- Schedule a flow with the Scheduler.
- Add SSH Client with Command
df -h. - Pass Message to a model that alerts on Slack if any disk goes over 80%.
Good practices
- Use a user with minimal permissions in the credential.
- If you give the component to an agent in tool mode, limit in its instructions which commands it may run: the agent runs what it writes.