← ClaudeAtlas

incidentlisted

Incident response and blameless postmortem — severity, containment steps, diagnosis (logs/metrics/containers via a deployment skill when present), fix/rollback, timeline, root cause, actions; writes docs/ops/incidents/INC-NNN.md.
gonimar/claude-web-studio · ★ 0 · DevOps & Infrastructure · score 76
Install: claude install-skill gonimar/claude-web-studio
# Incident Reply in the project conversation language (CLAUDE.md → Language); code, identifiers, paths and commit messages stay in English. Template `incident-postmortem.md`. ## Phase 1: Containment Severity; what users see; immediate measures (rollback via `/deploy`/the deployment skill, feature kill switch, rate limit) — with confirmation. Security (leak/breach?) → `security-lead` immediately: isolate, rotate secrets, preserve logs. ## Phase 2: Diagnosis Logs (delegate verb `logs [service] [--since]` by `docs/deploy-target-contract.md`, or `docker compose logs`), metrics, recent deploys/migrations, `git log`; hypotheses → verification. ## Phase 3: Fix `/hotfix` or rollback; verify by metrics. ## Phase 4: Postmortem Timeline, root cause (system/process, blameless), what worked/did not, actions (fix/prevent/detect) with owners and dates. "May I write `docs/ops/incidents/INC-NNN.md`?" — one `AskUserQuestion`: write (Recommended) · show the draft/diff first · not now. Actions → roadmap. After the "write" answer: `touch .claude/.write-consent` (rule 7 — the consent-guard hook checks the marker). Verdict: `RESOLVED` | `MITIGATED` | `OPEN`. Next step — one `AskUserQuestion`: stories for the prevent/detect actions (Recommended) · `/hotfix` · close the incident.