> For the complete documentation index, see [llms.txt](https://docs.gotempest.app/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.gotempest.app/productivity/snippets-scheduled-runs.md).

# SSH Snippets & Scheduled Commands

**Snippets** are Tempest's saved shell commands — paste a command once, give it a name, and re-run it on any host with a click. **Scheduled runs** turn snippets into cron jobs that fire on the hosts you choose, with notifications routed to your phone via [Tempest Push](/productivity/tempest-push-notifications.md) when a run fails.

It's the missing layer between "I keep retyping the same `journalctl -u …`" and "I don't want to set up Ansible just for three commands."

## Save a snippet

1. Open the **Snippets** panel.
2. Click **New Snippet**.
3. Paste your command (multi-line is fine — bash heredocs, pipelines, etc.).
4. Give it a name and optionally a description.
5. Save.

Snippets sync across your devices via your encrypted vault, just like hosts.

## Run a snippet

* In any SSH tab, open the snippet panel and pick one — Tempest sends it through the channel
* Or **target multiple hosts** at once: pick the snippet, select hosts from your list, run. Each host gets its own output panel.

## Schedule a snippet

1. Open a snippet → **Schedule** tab.
2. Set the cron expression (UI helper for common patterns: every hour, every day at 3 AM, etc.).
3. Pick targets: one host, a tag-matched group, or all hosts.
4. Optional: enable **Notify on failure** → exit code ≠ 0 triggers a Tempest Push to your phone.
5. Save.

The schedule runs **in the Tempest backend** (Electron main process / standalone server), not in the renderer. So:

* Your laptop can be asleep and the run still happens — as long as the Tempest backend is up.
* On the **standalone Tempest server**, schedules survive even if no client is connected (perfect for lightweight ops automation).

## Run history

Each scheduled run records:

* Start / end timestamp
* Exit code
* Stdout / stderr captured (configurable size limit)
* Per-target status when running fan-out

Browse history in **Snippets → History** with filters by snippet, host, and status.

## When to use snippets vs. an actual config-management tool

| Use Snippets when                                                       | Use Ansible / Puppet / Salt when                  |
| ----------------------------------------------------------------------- | ------------------------------------------------- |
| 1–10 hosts                                                              | Hundreds of hosts                                 |
| Operational commands ("show me the disk usage", "restart this service") | Declarative state ("nginx must be installed")     |
| You want one-shot fan-out from your laptop                              | You want a CI pipeline triggering converged state |
| Quick fix for a specific incident                                       | Reproducible infrastructure                       |

The two complement each other — snippets handle ad-hoc operational tasks, real config-management tools handle long-lived state.

## Pro feature

Scheduled runs are part of **Tempest Pro**. Saving and one-off running of snippets is free.

## See also

* [Tempest Push Notifications](/productivity/tempest-push-notifications.md) — how failure alerts reach your phone
* [Tempest AI Assistant](/productivity/tempest-ai-assistant.md) — generate snippet content from a natural-language description