Generate a Runbook Update Checklist After Every Incident

What is a Runbook Delta?

A runbook delta is a targeted list of changes your team should make to operational documentation — runbooks, playbooks, alerting guides, and escalation procedures — as a direct result of what you learned during an incident.

Unlike general action items, a runbook delta is scoped specifically to documentation: which file to update, what to add or change, and who owns each update. The goal is to make the next on-call engineer better prepared than the engineer who handled today's incident.

How AutoBrief Generates the Checklist

After you log an incident in AutoBrief, the runbook delta output analyzes your action items, root cause, and contributing factors to produce a structured checklist covering:

• Specific runbook sections to update, add, or rewrite
• Suggested additions to alerting and escalation thresholds
• Gaps in on-call documentation exposed during the incident
• New commands, procedures, or thresholds worth documenting
• Owner assignments for each documentation task

This checklist is designed to survive the post-incident handoff — concrete enough to assign as a ticket, not vague enough to defer indefinitely. You can also generate a full incident postmortem and a status page update from the same form.

Sample Runbook Delta Output

An abbreviated example from a database failover incident:

Runbook Delta — Database Failover Incident (2024-01-15)

Runbook: database-failover.md
  □ Add step: check replica lag before rerouting traffic
    Owner: SRE | Priority: High
  □ Add monitoring link for replication status dashboard
    Owner: SRE | Priority: Medium
  □ Update expected failover time estimate from 5 min to 10–15 min
    Owner: SRE lead | Priority: High

Playbook: batch-jobs.md
  □ Add rate limiting requirements for all batch queries
    Owner: Platform | Priority: High
  □ Define "off-peak" with explicit traffic thresholds, not time-of-day
    Owner: Platform lead | Priority: Medium

Alert Thresholds
  □ Create alert: replica_lag > 30s triggers SEV-3 page
    Owner: SRE | Priority: Critical
  □ Lower batch_query_duration alert from 60s to 30s
    Owner: Platform | Priority: Medium

On-Call Reference
  □ Add API gateway rollback command to quick-reference card
    Owner: SRE | Priority: Low

Operational documentation stays inside your team

Runbooks and playbooks often contain sensitive infrastructure details. AutoBrief stores all incident and output data in your private, encrypted workspace. No other organization can see your runbook gaps or documentation structure. Nothing is used to train AI models. Read more on our security page.