Prepare an incident post-mortem meeting agenda

Build a blameless post-mortem meeting plan that keeps engineering teams focused on systemic fixes instead of pointing fingers.

You getMeeting Objective & Ground Rules, timed Agenda and action Item Tracking Table

Prompt

Prompt
Act as a site reliability engineering lead. Ask me 3 short numbered questions, one by one or in a single message, to understand the incident: 1) What was the incident summary and user impact? 2) How long did it take to detect and mitigate? 3) What teams were involved in the response? Wait for my answers before writing the agenda.

Rules:
- Focus on learning, systemic safeguards, and timeline reconstruction, strictly avoiding individual blame.
- Allocate realistic time limits for a 50-minute meeting.
- Include specific prompts the facilitator should ask to identify root causes and latent systemic risks.

Output format:
1. Meeting Objective & Ground Rules
2. Timed Agenda (time block, topic, facilitator prompt)
3. Action Item Tracking Table (columns: Proposed Fix, Owner Role, Preventive Layer)
4. Facilitator Guardrails (what to say if participants become defensive)

Example output

Example for: checkout service outage lasting 45 minutes

  1. Meeting Objective & Ground Rules
  • Objective: Identify contributing systemic factors to the payment outage and commit to preventive hardening.
  • Ground Rules: Assume everyone made the best decisions possible given the information they had. Focus on process, tooling, and telemetry rather than individual mistakes.
  1. Timed Agenda
  • 00:00-00:05 | Context & Welcome: State the blameless agreement and review the impact scope.
  • 00:05-00:20 | Timeline Reconstruction: Walk minute-by-minute through detection, escalation, and rollback. Prompt: "Where did our monitoring give clear signals, and where were we operating blind?"
  • 00:20-00:35 | Contributing Factors & Safeguards: Analyze failure modes. Prompt: "What technical safeguards should have prevented this bad configuration from reaching production?"
  • 00:35-00:45 | Corrective Action Planning: Agree on concrete tasks. Prompt: "Which single change reduces the blast radius most effectively?"
  • 00:45-00:50 | Next Steps: Confirm action item owners and review dates.
  1. Action Item Tracking Table
Proposed FixOwner RolePreventive Layer
Add schema validation check on payment deploymentRelease EngineerPre-deployment gate
Add synthetic ping for zero-success payment transactionsSRE LeadAlerting and telemetry
Update rollback playbook for configuration pushesOn-call PrimaryMitigation velocity
  1. Facilitator Guardrails
  • If someone says: "Alex should have spotted the syntax error during the review."
  • Facilitator redirect: "Our review process should catch invalid syntax automatically before human eyes review it. How do we automate that lint check?"
Written by AI from this prompt. Your results will vary.

Run this prompt in ChatGPT, Claude or Gemini and reply to the 3 questions with raw incident details. Ask the model to adjust the time blocks if you only have a 30-minute meeting window.

Make it repeatable

Keep using this prompt without pasting it in every time.

Keep exploring

More in Meetings & Notes