Skills incident-hotfix
Coder-focused incident response and hotfix execution for production issues. Use when you need reproducible triage, patch/rollback decisions, CI-safe hotfix branches, evidence capture, and postmortem action tracking tied to code changes.
install
source · Clone the upstream repo
git clone https://github.com/openclaw/skills
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/openclaw/skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/broedkrummen/incident-hotfix" ~/.claude/skills/clawdbot-skills-incident-hotfix && rm -rf "$T"
manifest:
skills/broedkrummen/incident-hotfix/SKILL.mdsource content
Incident Hotfix
Use this alongside broader incident-response skills when code-level action is required.
Workflow
- Classify severity using
references/severity-matrix.md - Create hotfix branch from current production tag/commit
- Reproduce and isolate with minimal failing test
- Patch with rollback plan
- Run focused CI checks
- Capture evidence bundle
- Merge + verify + postmortem actions
Step 1 — Create incident workspace
bash scripts/start_hotfix.sh --id INC-1234 --base main
This creates:
branchhotfix/INC-1234-<slug>
folderdocs/incidents/INC-1234/- starter files for timeline, rollback, and actions
Step 2 — Evidence capture
bash scripts/capture_evidence.sh --id INC-1234
Collects into
docs/incidents/INC-1234/evidence/:
- latest commits + diff summary
- changed files list
- local env snapshot (safe subset)
- test output logs
Step 3 — Patch gate
Before PR/merge, ensure:
- failing case reproduced (or clearly documented)
- minimal patch scope
- rollback command documented in
ROLLBACK.md - focused tests pass + no new lint/type failures
Step 4 — Postmortem actions
Use
references/action-template.md to convert findings into concrete tasks:
- owner
- due date
- verification criteria
Required outputs
docs/incidents/<id>/TIMELINE.mddocs/incidents/<id>/ROLLBACK.mddocs/incidents/<id>/ACTIONS.md
bundledocs/incidents/<id>/evidence/
Notes
- Prefer smallest safe patch over broad refactor during incident.
- If root cause is uncertain, ship containment first, then permanent fix.
- Never merge hotfix without rollback path.