hotfix

Fast-track fixes for production incidents.

When to Use

• Production is down or degraded
• Critical bug affecting users
• Security vulnerability discovered

How It Differs from Normal Flow

| Normal | Hotfix | |--------|--------| | Branch from develop | Branch from main | | ASK mode exploration | Skip to fix | | Full PR review | Expedited review | | Standard Slack message | URGENT alert |

The Flow

1. Branch - Create hotfix branch from main
2. Fix - Implement minimal fix
3. Verify - Test fix works
4. PR - Create with HOTFIX label
5. Alert - Notify team urgently

Phase Checklist

Phase 1: Create Hotfix Branch

• [ ] Fetch latest: git fetch origin main
• [ ] Branch from main: git checkout -b hotfix/<issue-summary> origin/main
• [ ] Confirm on hotfix branch

Phase 2: Implement Fix

• [ ] Identify root cause (minimal investigation)
• [ ] Implement smallest possible fix
• [ ] Run typecheck and lint
• [ ] NO new features, NO refactoring

Phase 3: Verify

• [ ] Test fix locally
• [ ] Confirm original issue resolved
• [ ] Check for obvious regressions

Phase 4: Create PR

• [ ] Push branch: git push -u origin HEAD
• [ ] Create PR to main with HOTFIX label:

  gh pr create --base main --title "HOTFIX: <summary>" --label hotfix
  

• [ ] Request expedited review

Phase 5: Alert Team

Generate urgent Slack message:

:rotating_light: **HOTFIX: [Issue Summary]**

**Impact:** [What was broken]
**Fix:** [What was changed]

:github: PR: [link]
:notion: Incident: [link if exists]

Requesting expedited review from: @[reviewer]

After Merge

• [ ] Merge main back to develop: git checkout develop && git merge main
• [ ] Create incident report in Notion (if major)