Documentation Grafana Cloud AI and machine learning Grafana Assistant Assistant rules Guide workflow escalation

Guide workflow escalation

Create a workflow guideline rule when you need Grafana Assistant to recommend operational follow-ups automatically. The rule ensures the Assistant prompts incident declarations, escalations, or deployment reviews whenever conversations match your criteria.

Before you begin

• Access the settings page: Open Grafana Assistant > Settings > Assistant behavior.
• Runbook alignment: Identify the triggers and escalation paths you expect the Assistant to reference.
• Audience: Confirm who should receive the guidance and whether it applies tenant-wide.

Launch the workflow guideline rule

1. In Assistant behavior, expand Customize your Assistant experience through behavioral rules.
2. Click Create rule, or select Add and edit this rule on the Workflow Guidelines card if it appears.

Write the escalation language

1. Name the rule, for example, Workflow Guidelines.

2. Describe triggers and recommended actions in the body:

   text Copy

   Suggest declaring an incident if the ecommerce platform response time exceeds 5 seconds. Recommend escalating to the infrastructure team for persistent memory issues. Always ask about recent deployments when troubleshooting performance issues.

3. Add any other signals or decision points relevant to your environment, such as paging rotations or affected regions.

Select scope and enablement

1. Choose Just me for personal experimentation or Everybody for org-wide automation.
2. Keep the rule enabled so the Assistant recommends these follow-ups during relevant conversations.

Save and test

1. Click Create rule or Save changes for template edits.
2. Start a fresh conversation that mimics a trigger scenario, such as discussing slow e-commerce response times, and check that the Assistant recommends the correct escalation steps.
3. Tune the rule text if the Assistant misses key conditions or recommendations.

Make workflow prompts automatic

Workflow escalation prompts now appear automatically, helping your team follow consistent incident management and troubleshooting paths without retyping runbook reminders.

Next steps

• Combine with Provide infrastructure and application hints to reference specific telemetry in escalations.
• Review Manage your data privacy and security to understand data handling, third-party providers, security measures, and access controls.