# Automation without a scripting degree

Source: https://iskue.com/features/automation/

> Triggers, conditions, actions — rules that respect your workflow.

Automation rules watch ticket activity in your project and act on it. When a trigger fires and every condition matches, the rule's actions run — after the triggering change commits, in a separate transaction, with each outcome logged to a run history. Rules are built from plain parts: one trigger or schedule, zero or more conditions, one or more actions.

## What you get

- **Triggers for every kind of work**: ticket created, updated, transitioned, assigned, commented or labelled — plus customer and public replies, request-type and field changes, SLA thresholds, approvals, a TQL match, an incoming webhook or a cron schedule.
- **Conditions that read like queries**: narrow the match by type, priority, status, label, a due date within N days, a custom field value — or any TQL query; a rule applies only when all of its conditions hold.
- **Actions that reach out**: assign, set priority, transition, add a label, comment or canned reply — and edit any field, create a ticket, send an email or call a webhook.
- **Runs after commit**: so a rule sees the ticket as it was actually saved; execution happens in its own transaction and never blocks or breaks the change that triggered it.
- **Workflow-validated transitions**: mean a transition action only moves a ticket along a path the project's workflow allows for that ticket type; anything else is skipped, not forced.
- **A run history**: records each matching run as applied, skipped, or failed — with an error message on failure — so your admins can review a rule's recent activity.

## Chains with a brake

Changes a rule makes are published like any other change, so webhooks, notifications and live boards see them — and one rule can trigger another. A rule never triggers itself, and a chain stops after three firings, so a misconfiguration can never spin into a loop. If one action in a rule fails, the remaining actions still run, and the failure is logged rather than swallowed.

## Scheduled and global rules

A rule can run on a clock instead of an event: hourly or daily, useful with the due-within condition for deadline sweeps. On multi-node deployments each due run is claimed with a compare-and-set update, so exactly one backend executes it. Org admins can also define global rules that apply across every project, using a deliberately restricted vocabulary — type, priority, and due-within conditions; assign, set-priority, and add-comment actions — with runs logged under each affected project.

## Attributed and audited

Assignments, priority changes, transitions and comments made by a rule all land in the ticket's change history. Comments added by a rule default to internal notes; on service-desk tickets a rule can mark a comment customer-visible, which posts it as a public reply and emails the requester. Creating, updating, or deleting a rule is itself recorded in the configuration audit log.

## Key facts

- 16 trigger kinds
- 7 condition fields
- Email and webhook actions
- Chains stop after 3 firings

## In practice

```
POST /api/v1/projects/{projectKey}/automation-rules
{
  "name": "Triage bugs",
  "enabled": true,
  "trigger": "CREATED",
  "conditions": [{ "field": "TYPE", "value": "BUG" }],
  "actions": [{ "type": "SET_PRIORITY", "value": "HIGH" }]
}
```
