メニュー
Write a blameless incident postmortem — timeline, impact, root cause, and concrete follow-ups, focused on systems not people.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | blameless-postmortem |
| description | Write a blameless incident postmortem — timeline, impact, root cause, and concrete follow-ups, focused on systems not people. |
| category | Documentation |
| tags | incident, postmortem, sre, writing |
After any incident worth learning from (outage, data issue, security event).
Blame the system, not the person. People act reasonably given the information and tools they had. If a single human error caused an outage, the system let it.
Write clear, conventional git commit messages — type(scope): summary, imperative mood, a body that explains the why.
Build small, secure container images — pinned base, non-root user, multi-stage builds, no secrets in layers, minimal attack surface.
Control and query Home Assistant through the connector's call_service tool — REST API patterns, common domains, how to find entities before acting.
Design predictable REST APIs — noun resources, correct status codes, pagination, consistent errors, and versioning.
Handle API keys, tokens and passwords safely — store them in the vault, reference by name, never in chat, repo, memory, or code.
Diagnose slow queries and add the right indexes — read the query plan, index for filters/joins/sorts, avoid over-indexing.
