菜单
This skill should be used when the user asks to "verify critic independence", "check for overlapping failures", "validate swiss cheese model", "are the critics independent", "measure orthogonality", or needs to verify that multiple critic layers have misaligned failure modes as required by the swiss cheese model.
This skill should be used when the user asks to "define acceptance criteria", "what are the success criteria", "set quality gates", "establish acceptance tests", "define what success looks like", or needs to specify pre-declared success criteria before code execution begins.
This skill should be used when the user asks to "show audit trail", "why did we make this decision", "what happened after this", "trace this decision", "show decision history", "search audit log", or needs to understand the decision history and rationale behind code changes verified through coherence check.
This skill should be used when the user asks to "check this code", "validate this", "verify this implementation", "is this correct", "review this code", "check for errors", "multi-agent verification", or mentions production-critical code, financial calculations, security implementations, or high-stakes operations. Provides comprehensive multi-agent verification workflow with specialized critic agents.
This skill should be used when the user asks to "review the plan", "check the execution plan", "validate this plan", "is this plan good", or automatically after the planner agent creates an execution plan. Validates plans for completeness, clarity, measurability, and proper acceptance criteria before execution begins.
This skill should be used when the user asks to "use backward design", "create curriculum using UbD", "design learning outcomes first", "start with desired results", "plan assessments before activities", "design assessments first", "create outcomes-based curriculum", or references Understanding by Design framework. Also applies when user mentions aligning outcomes with assessments or avoiding activity-first planning. Provides comprehensive guidance on applying backward design methodology to create effective curricula for 1-2 day intensive workshops.
This skill should be used when the user asks to "write learning objectives", "create measurable learning outcomes", "select action verbs for objectives", "what cognitive levels should I target", "how to scaffold learning", "write course objectives", "make objectives more specific", or references Bloom's taxonomy. Also relevant when users say "my objectives aren't measurable", "how do I assess what students learned", or "progression from basic to advanced skills". Provides comprehensive guidance on applying Bloom's taxonomy to create measurable, appropriately-leveled learning objectives for intensive workshops.
| name | Swiss Cheese Validation |
| description | This skill should be used when the user asks to "verify critic independence", "check for overlapping failures", "validate swiss cheese model", "are the critics independent", "measure orthogonality", or needs to verify that multiple critic layers have misaligned failure modes as required by the swiss cheese model. |
Verify that multiple critic layers have orthogonal failure modes, ensuring the Swiss cheese model is working correctly. Measure overlap in critic rejections and warn if critics catch the same errors.
The Swiss cheese model achieves reliability through misaligned holes:
Target: Independence score > 0.85 (less than 15% overlap)
Analyze rejection patterns across sessions:
code_rejections = sessions_where_code_critic_vetoed
security_rejections = sessions_where_security_critic_vetoed
overlap = code_rejections ∩ security_rejections
union = code_rejections ∪ security_rejections
independence_score = 1 - (|overlap| / |union|)
Interpretation:
From research paper (522 sessions):
| Critic Pair | Overlap | Independence |
|---|---|---|
| Code vs Security | 2.3% | 0.977 ✓ |
| Code vs Domain | 3.1% | 0.969 ✓ |
| Security vs Domain | 0.8% | 0.992 ✓ |
High overlap detected when:
Causes:
Solutions:
Default (hardcoded):
layers:
- code-critic
- security-critic
- domain-critic
Custom (user-configured):
layers:
- code-critic
- security-critic
- financial-critic
- performance-critic
Dynamic (context-based):
layers:
financial_files: [code-critic, financial-critic]
auth_files: [code-critic, security-critic]
api_files: [code-critic, performance-critic]
SWISS CHEESE VALIDATION REPORT
Analysis Period: Last 30 days (87 sessions)
INDEPENDENCE SCORES:
✓ Code vs Security: 0.981 (excellent)
✓ Code vs Domain: 0.973 (excellent)
✓ Security vs Domain: 0.995 (excellent)
OVERLAP ANALYSIS:
- Code & Security both rejected: 2 sessions (2.3%)
Reasons: Authentication logic had both syntax error AND security flaw
- Code & Domain both rejected: 3 sessions (3.4%)
Reasons: Financial calculation had logic error AND violated precision rules
- Security & Domain both rejected: 1 session (1.1%)
Reasons: Data validation missing AND violated GDPR requirements
CONCLUSION: ✓ Swiss cheese model working correctly
All critics demonstrate orthogonal failure modes.
⚠ INDEPENDENCE CONCERN DETECTED
Analysis: Last 30 days (87 sessions)
POOR INDEPENDENCE:
❌ Code vs Security: 0.72 (poor - 28% overlap)
ISSUE: Both critics rejecting same sessions frequently
Sample overlaps:
- Session #42: Both caught missing input validation
- Session #55: Both caught SQL injection in user input
- Session #61: Both caught unescaped HTML in output
ROOT CAUSE: Security critic checking code-level details
RECOMMENDATION: Refocus security critic on OWASP patterns only,
let code critic handle general logic validation
Run validation automatically after every 20 sessions:
monitoring:
frequency: every_20_sessions
alert_threshold: 0.80
action: warn_user
references/independence-verification.md - Statistical methods for measuring orthogonalityreferences/failure-mode-analysis.md - Common causes of critic overlapexamples/poor-independence.md - Case study of overlapping critics and remediation