Skip to main content

How Automation Creates Action Items

Action items—tasks and follow-ups—can be generated automatically through Workflows, Incident Roles, or the Rootly API. Automation ensures teams never miss a critical task, follow-up, or improvement opportunity during or after an incident. Automation is especially powerful for:
  • Enforcing consistent response processes
  • Creating tasks or follow-ups automatically based on incident conditions
  • Routing work to the correct owners
  • Producing reliable audit trails
  • Integrating with tools such as Jira, GitHub, and Slack
Automation reduces operational overhead and ensures every incident produces actionable, trackable work.

Action Items and Workflows

Workflows can both create and react to action items using a variety of triggers and conditions.

Supported Workflow Triggers (Action Item Category)

Rootly supports the following action-item–related triggers:
  • incident_updated
  • action_item_created
  • action_item_updated
  • assigned_user_updated
  • summary_updated
  • description_updated
  • status_updated
  • priority_updated
  • due_date_updated
  • teams_updated
  • slack_command

Workflow Conditions

Workflows can filter based on:
  • Action item type (Task / Follow-up)
  • Status
  • Priority
  • Incident severity
  • Visibility
  • Incident kind
  • Incident roles
  • Other incident attributes (teams, services, etc.)
Conditions and triggers map directly to code-backed enums and workflow schemas, ensuring strict validation and predictable automation.

Example: Workflow creates a task when the Security team is added

Trigger
  • Teams added
Conditions
  • Kind → Incident
  • Team → is one of → Security
Action
  • Create a task to alert the Legal team

Example: Workflow reacts to a new action item

Trigger
  • Action item created
Conditions
  • Type → Task
  • Priority → High
Action
  • Create an external ticket (e.g., Jira, GitHub, GitLab, Linear)
Workflow tasks are grouped by integration. Jira actions, for example, appear under the Jira task group and require the Jira integration to be enabled.

Action Items and Incident Roles

Incident Roles can include predefined tasks that are automatically converted into incident action items when an incident is created. These role-based tasks carry:
  • Summary
  • Priority
  • Role assignment metadata
  • Ordering/position

To configure:

  1. Go to Configuration → Roles
  2. Select a role
  3. Open the Tasks tab
  4. Add or reorder tasks
Role-based action items give each incident a predictable starting checklist and ensure operational discipline.
Learn more about Incident Roles.

Action Items and the API

The Rootly API allows programmatic creation, management, and retrieval of action items.

Endpoints

Supported API Fields (Create/Update)

  • kind (task or follow_up)
  • summary (required)
  • description (Markdown supported)
  • assigned_to_user_id
  • assigned_to_group_ids
  • priority (high, medium, low)
  • status (open, in_progress, done, cancelled)
  • due_date (ISO 8601)
  • Jira fields:
    • jira_issue_id
    • jira_issue_key
    • jira_issue_url

Response Fields Include

  • Kind, priority, status
  • Due date
  • Assigned user & groups
  • Integration URLs:
    • jira_issue_url
    • github_issue_url
    • gitlab_issue_url
    • linear_issue_url
  • url and short_url
  • Timestamps and metadata
The API follows the JSON:API spec and enforces the same validations as the Web UI and Slack.

Best Practices

  • Automate common tasks to reduce manual work
  • Use role-based tasks for consistent incident startup actions
  • Assign owners early to prevent drift
  • Use priorities intentionally to structure follow-up workflows
  • Integrate external systems (Jira, GitHub, etc.) for centralized tracking
  • Review overdue follow-ups regularly for reliability improvements

Troubleshooting

Confirm the trigger conditions matched the incident and that the user/action had permission to create action items.
Ensure the role is enabled and the role tasks themselves are active. Disabled tasks are not copied over.
Make sure summary is present and enum fields (priority, status, kind) match allowed values. Check Jira fields if provided.
Some organizations disable task/follow-up creation after an incident is resolved, cancelled, or closed.
Slack notifications fire only when:
• The team has a Slack integration

• The incident has a slack_summary_timestamp

• Notifications aren’t suppressed

These notifications are not workflow-dependent.