Generates a structured standard operating procedure for incident response, covering detection, escalation, communication templates by severity level, and a triage decision tree.
Generates a structured standard operating procedure for incident response, covering detection, escalation, communication templates by severity level, and a triage decision tree. Use it when your team needs a repeatable playbook for handling production failures or operational disruptions.
Create an SOP for handling an exception or incident.
SITUATION: [what goes wrong]
SEVERITY: [how bad is it — annoying vs. critical]
FREQUENCY: [how often this happens]
CURRENT PROCESS (rough):
[How you handle this today — even if it's ad hoc]
The SOP should include:
1. DETECTION: How do you know this has happened? (alerts, customer reports, monitoring)
2. TRIAGE: How to assess severity and who to notify
3. IMMEDIATE RESPONSE: First 15 minutes — stop the bleeding
4. INVESTIGATION: How to find the root cause
5. RESOLUTION: Steps to fix it, organized by common causes
6. COMMUNICATION: Templates for customer communication at each severity level
7. POST-MORTEM: What to document after resolution
8. PREVENTION: Steps to reduce future occurrences
Include a decision tree for triage — the first thing someone should look at to determine which resolution path to follow.
Source: https://craighewitt.com/claude-playbook/operations/sop-machine