Deployment Validator Skill

Comprehensive pre-deployment validation to ensure production readiness and minimize deployment risk.

When This Skill Activates

• Before deploying any code to production
• Before finalizing a new schedule for deployment
• Before applying database migrations
• Before infrastructure changes
• During emergency hotfix deployment

Validation Framework

Pre-Deployment Checklist

Code Quality Gate

• [ ] All tests passing (unit, integration, e2e)
• [ ] Type checking passes
• [ ] Linting passes
• [ ] Code coverage requirements met
• [ ] Security audit completed
• [ ] No critical/blocker issues

Database Migration Gate

• [ ] Migration tested on staging
• [ ] Rollback procedure documented
• [ ] Data backup created
• [ ] Migration time estimated
• [ ] Backward compatibility verified

Schedule Deployment Gate

• [ ] Schedule validated for ACGME compliance
• [ ] Coverage gaps resolved
• [ ] Faculty approvals obtained
• [ ] Resident notifications ready
• [ ] Contingency plan documented

Infrastructure Gate

• [ ] Capacity verified
• [ ] Monitoring configured
• [ ] Alerting rules active
• [ ] Rollback plan documented
• [ ] Health checks configured

Phase 1: Code and Quality Validation

1. Check test results
   - Unit tests: must pass
   - Integration tests: must pass
   - Coverage: must exceed threshold

2. Run security checks
   - OWASP Top 10 scan
   - Dependency vulnerability check
   - Secrets detection

3. Verify code quality
   - Linting: must pass
   - Type checking: must pass
   - Complexity: within limits

Phase 2: Database Readiness

1. Migration validation
   - Syntactically correct
   - No unsafe operations
   - Tested on copy of prod data
   - Rollback working

2. Data integrity
   - Backup created
   - Constraints still satisfied
   - Foreign keys valid
   - Indexes updated

3. Performance
   - Migration time acceptable
   - No long locks
   - Monitoring in place

Phase 3: Schedule Validation

1. Compliance check
   - ACGME rules verified
   - Coverage adequate
   - Staffing realistic

2. Operational check
   - Faculty confirmed availability
   - Rotation timing valid
   - Contingencies in place

3. Communication
   - Residents notified
   - Faculty acknowledged
   - Adjustments documented

Phase 4: Infrastructure and Monitoring

1. Capacity verification
   - CPU/memory adequate
   - Database connections sufficient
   - Network bandwidth available

2. Monitoring setup
   - Metrics collection active
   - Alerting rules configured
   - Dashboards ready
   - Health checks in place

3. Incident response
   - On-call team ready
   - Rollback procedure documented
   - Communication channels ready

Risk Assessment Matrix

| Risk Factor | Low | Medium | High | |-------------|-----|--------|------| | Code changes | <100 lines | 100-500 lines | >500 lines | | Test coverage | >90% | 80-90% | <80% | | Database changes | Add column | Schema restructure | Table drop | | Deployment scope | Single component | Multiple components | System-wide | | Rollback time | <5 minutes | 5-30 minutes | >30 minutes |

Risk Score = Sum of risk levels

• Low risk (0-5): Proceed with review
• Medium risk (6-12): Require additional testing
• High risk (13+): Escalate to human decision

Deployment Validation Report

## Deployment Readiness Report

**Release:** [VERSION]
**Date:** [DATETIME]
**Deployment Type:** [CODE/DATABASE/SCHEDULE/INFRA]

### Risk Assessment
- Overall Risk Level: [LOW/MEDIUM/HIGH]
- Risk Score: [N/10]
- Blockers: [COUNT]

### Quality Gates
- [x] Code quality
- [x] Tests passing
- [x] Security audit
- [x] Performance acceptable
- [x] Monitoring ready

### Deployment Checklist
- [ ] Pre-deployment steps complete
- [ ] Rollback procedure documented
- [ ] Team notification sent
- [ ] Health checks configured
- [ ] On-call team briefed

### Critical Items
[List anything requiring attention before deployment]

### Recommendation
- [APPROVED / CONDITIONAL / BLOCKED]

### Next Steps
1. [Action 1]
2. [Action 2]

Quick Validation Commands

# Full deployment validation
python -m app.deployment.validator --release=current --full

# Skip tests (if already run)
python -m app.deployment.validator --release=current --skip-tests

# Database migration check
python -m app.deployment.validator --type=migration --path=alembic/versions/xxx

# Schedule deployment validation
python -m app.deployment.validator --type=schedule --schedule_id=current

Common Deployment Scenarios

Scenario 1: Hotfix Deployment

Risk: Medium (time pressure + changes) Validation:

• Tests for hotfix passes
• No regression in related tests
• Rollback tested
• Communication plan ready

Scenario 2: Major Feature Release

Risk: High (large change scope) Validation:

• Full test suite passes
• Staged rollout plan
• Monitoring for metrics
• Communication multi-channel

Scenario 3: Database Migration

Risk: Variable (depends on scope) Validation:

• Migration tested on staging
• Data backup confirmed
• Rollback tested
• Downtime window approved

Scenario 4: Schedule Deployment

Risk: Medium (operational impact) Validation:

• ACGME compliant
• Coverage verified
• Staffing confirmed
• Contingency documented

Escalation Decision Tree

Are there deployment blockers?
├─ YES → BLOCK deployment
│  └─ Escalate to human decision
└─ NO → Continue

Are there critical warnings?
├─ YES → Require human approval
│  └─ Document reasoning
└─ NO → Continue

Is rollback time acceptable?
├─ NO → Require automated rollback setup
└─ YES → Proceed

Is on-call team ready?
├─ NO → Delay deployment
└─ YES → APPROVED for deployment

References

• See PROMPT_LIBRARY.md for deployment validation templates
• See CLAUDE.md for deployment procedures
• Incident response procedures in incident-responder skill