Automations

Automations are scheduled agent runs. Instead of you typing a prompt, NimbleBrain runs a saved prompt on a schedule — a daily digest, a recurring report, a periodic check — and the agent executes it unattended, with access to your tools and connectors.

Each automation is a prompt plus a schedule. When it fires, the runtime starts an agent run framed around producing a deliverable rather than a back-and-forth conversation.

What an automation is

An automation has:

• A name — becomes a kebab-case ID (for example, “Weekly Report” → weekly-report)
• A prompt — the instruction sent to the agent on each run
• A schedule — either a cron expression or a fixed interval
• Optional run-time policy — a model override, a forced skill match, iteration and token caps

Who an automation runs as

Automations are owned by the user who creates them, not by a workspace. A scheduled run fires as its owner, focused on the workspace it was created in, but with full reach across every workspace that owner belongs to.

This has one practical consequence: if an automation needs to reach a connector (for example, to post to Teams or Slack), that connector must be installed in a shared workspace the owner is a member of. A connector that lives only in someone’s personal workspace is never reachable by another user’s automation.

Note

Automation runs use your timezone for cron scheduling. The default is Pacific/Honolulu, overridable by the operator with the NB_TIMEZONE environment variable. A schedule can also carry its own IANA timezone.

Schedules

Two schedule types:

• Cron — a standard 5-field cron expression (for example, 0 9 * * 1 for 9:00 AM every Monday)
• Interval — a fixed interval in milliseconds, minimum 60000 (one minute)

Creating an automation

The simplest way is to ask the agent in chat. Describe what you want and when:

Every weekday at 8am, summarize my unread email and post it to the
#daily channel.

The agent creates the automation for you, generating the schedule and the prompt. You can ask it to list, pause, resume, or delete automations the same way.

Managing automations

Browse and manage automations from the Automations panel in the sidebar, or ask the agent in chat. Both surfaces let you:

1. List automations — name, enabled state, schedule, source, last run status, and next run time.

2. Inspect one automation — its full definition (schedule, model, caps, prompt preview) and recent run history with per-run status, start time, iterations, and tool-call counts.

3. Pause an automation — disable it without deleting.

4. Resume a paused automation — re-enable it on its existing schedule.

5. Trigger an immediate run — bypass the schedule and run now. Short runs return the full run record; runs that take longer than about 30 seconds continue in the background, and the panel updates when they finish.

Tip

Pausing keeps the definition and its run history intact; the scheduler simply skips it. Resume re-enables it on its existing schedule. To remove an automation entirely, ask the agent to delete it (run history is preserved even then).

Run history

Every run is recorded with its status, start and completion times, iteration count, tool calls, and any error. View it in the Automations panel or by asking the agent. If the runtime restarts while a run is in flight, the orphaned run is swept and marked failed so history stays accurate.

What’s next

• Chat — the interactive counterpart to a scheduled run
• Connectors — install the connectors an automation posts to in a shared workspace
• Workspaces — owner reach and shared-workspace semantics