Skip to main content

Postmortem: [Incident Title]

Date: YYYY-MM-DD
Author: [Member or human who authored this]
Status: Draft | Review | Final

Incident Summary

<!-- One paragraph. What happened, when, and what systems or users were affected. -->

Timeline

Time (UTC)Event
HH:MM[Detection — who or what first noticed]
HH:MM[Response — who was paged, what action was taken]
HH:MM[Mitigation — how impact was reduced or stopped]
HH:MM[Resolution — when normal operation was restored]

Root Cause

<!-- Produced by The Debugger. What was the underlying defect or condition? Include: - The chain of events from root cause to visible symptom - Why existing tests or checks did not catch it - The specific file, function, or configuration at fault -->

Impact

<!-- Produced by The Tester. What broke and for whom? Quantify where possible: - Number of users affected - Duration of impact - Error rates, latency degradation, or data loss -->

Remediation

<!-- What was done to fix the immediate issue. Include PR/commit references. -->
  • [PR # or commit] — [short description]

Prevention

<!-- Informed by The Auditor. What systemic changes prevent recurrence? Examples: new tests, additional validation, monitoring, process changes. -->
  • [Action item with owner]
  • [Action item with owner]

Decision Record

<!-- Link to the corresponding Decision Log entry, if one exists. -->

See Decision Log entry and its provenance entry. Run ProvenanceStore.verifyChain() to prove the trail is intact.

Related

<!-- Links to relevant issues, ADRs, or documentation. -->
  • Issue #[number]
  • ADR-[NNN]
  • [Other references]

Generated by the Agenthood Society. Consumed by The Librarian to feed test cases, standards, and checklists.