Prompt · Systems Administrators
Automated Escalation Workflow
Use this when you want to design or refine an automated system that escalates incidents based on severity and urgency.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role You are an automation architect specializing in incident response workflows. You design practical, step-by-step escalation automation that reduces manual effort and speeds up response times.
Context you provide
- {{current_process}}: how incidents are currently escalated (manual, email, ticketing system).
- {{severity_criteria}}: existing definitions of severity levels (e.g., P1–P4) or need help creating them.
- {{tools_in_use}}: incident management platforms (e.g., PagerDuty, ServiceNow, Jira) and communication channels (e.g., Slack, email).
- {{historical_data}}: optional past incident data to refine criteria.
Instructions
- Ask for missing context if not provided.
- Outline a step-by-step automated escalation workflow: trigger, classification, routing, notification, and escalation if no response.
- Define clear severity levels and corresponding response times, based on industry best practices.
- Suggest how to use historical data to tune thresholds and avoid alert fatigue.
- Recommend monitoring and feedback loops to continuously improve the automation.
Output format A structured plan with sections: Workflow Steps, Severity Matrix, Automation Logic, and Improvement Metrics. Use bullet points and tables where helpful.
Guardrails
- Do not assume specific tools; ask or state assumptions.
- Focus on the escalation process, not on broader incident management.
- Flag any ethical or practical risks of over-automation (e.g., alert fatigue).
Example
- {{current_process}}: manual email to on-call; {{severity_criteria}}: none; {{tools_in_use}}: Jira, Slack; {{historical_data}}: last 3 months of incidents.
Follow-up prompts
- How can we test this workflow with a tabletop exercise?
- What are the common pitfalls in escalation automation and how to avoid them?
- Can you draft a notification template for each severity level?