assumption-testing（假设检验）

Assumption Testing

Identify, document, and explicitly test implicit assumptions in code, tests, design, and documentation to prevent failures caused by untested beliefs about how the system should work.

When to use me

Use this skill when:

• Building new features with uncertain requirements
• Working with legacy code or undocumented systems
• Integrating with external systems or third-party services
• Designing complex algorithms or business logic
• Testing edge cases and boundary conditions
• Preparing for production deployment of critical features
• Debugging mysterious or intermittent failures
• Reviewing code or architecture designs
• Onboarding to a new codebase or system

What I do

1. Assumption Discovery & Documentation

• Scan code for implicit assumptions:

- Data format and validation assumptions - Dependency availability and behavior assumptions - Performance and resource availability assumptions - User behavior and interaction assumptions - Error handling and recovery assumptions - Time and timing-related assumptions - Security and authorization assumptions

• Analyze tests for hidden assumptions:

- Test data assumptions (representative? comprehensive?) - Test environment assumptions (matches production?) - Test execution order assumptions (independent?) - Test assertion assumptions (complete? accurate?) - Test coverage assumptions (what's actually tested?)

• Review design and documentation:

- Architectural assumptions (scalability, availability) - Integration assumptions (APIs, protocols, formats) - User experience assumptions (workflows, expectations) - Business logic assumptions (rules, calculations) - Operational assumptions (monitoring, maintenance)

2. Assumption Categorization & Prioritization

• Categorize by risk and impact:

- Critical assumptions: Failure causes system failure - High-impact assumptions: Failure causes major issues - Medium-impact assumptions: Failure causes noticeable issues - Low-impact assumptions: Failure causes minor issues

• Prioritize based on:

- Likelihood of assumption being wrong - Impact if assumption is wrong - Cost to test the assumption - Age and staleness of the assumption - Evidence supporting the assumption

3. Assumption Testing & Validation

• Design explicit assumption tests:

- Boundary tests: What if assumption boundaries are exceeded? - Negation tests: What if the assumption is false? - Variation tests: What if assumption parameters vary? - Stress tests: What if assumption limits are pushed? - Reality checks: Does assumption match real-world behavior?

• Execute assumption validation:

- Automated tests for testable assumptions - Manual verification for complex assumptions - Simulation and modeling for theoretical assumptions - User research for behavioral assumptions - Monitoring and observation for operational assumptions

4. Assumption Management & Communication

• Document assumptions explicitly:

- Create assumption registry or catalog - Link assumptions to code and tests - Track assumption validation status - Document assumption sources and rationale

• Communicate assumption risks:

- Share assumption catalog with team - Highlight critical untested assumptions - Recommend assumption testing priorities - Report assumption validation results

Common Assumption Types to Test

Technical Assumptions:

• "The database will always be available"
• "API responses will match the documented schema"
• "Network latency will never exceed 100ms"
• "Memory will always be sufficient for operations"
• "Third-party services will always respond within SLA"

Business Logic Assumptions:

• "Users will always enter valid data"
• "Currency conversion rates will be available"
• "Business hours are 9-5 Monday-Friday"
• "Tax calculations follow current regulations"
• "Inventory levels are accurately tracked"

User Experience Assumptions:

• "Users understand how to use this feature"
• "Mobile users have stable internet connections"
• "Users will notice and read error messages"
• "Users prefer this workflow over alternatives"
• "Accessibility requirements are being met"

Operational Assumptions:

• "Monitoring will alert us of failures"
• "Backups are complete and restorable"
• "Scaling will handle peak traffic"
• "Security patches are applied promptly"
• "Documentation matches current implementation"

Examples

# Discover assumptions in code
npm run assumption:discover -- --path src/payment/
npm run assumption:discover -- --type technical
npm run assumption:discover -- --type business-logic

# Document and categorize assumptions
npm run assumption:catalog -- --output assumptions.json
npm run assumption:prioritize -- --risk high

# Test specific assumptions
npm run assumption:test -- --assumption "database-always-available"
npm run assumption:test -- --assumption "api-schema-matches-docs"
npm run assumption:test -- --assumption "user-enters-valid-data"

# Comprehensive assumption testing
npm run assumption:test:all              # Test all documented assumptions
npm run assumption:test:critical         # Test only critical assumptions
npm run assumption:test:untested         # Test assumptions without validation

# Integration with other testing
npm run assumption:test:with -- --test-type chaos --assumption "system-resilient"
npm run assumption:test:with -- --test-type performance --assumption "scales-linearly"
npm run assumption:test:with -- --test-type security --assumption "no-unauthorized-access"

# Assumption management
npm run assumption:track -- --status validated     # Show validated assumptions
npm run assumption:track -- --status untested      # Show untested assumptions
npm run assumption:track -- --status risky         # Show high-risk assumptions

Output format

Assumption Testing Report
──────────────────────────────
Scope: Payment Processing Module
Assumptions Discovered: 28
Assumptions Tested: 15 (priority order)
Testing Duration: 3 hours

Critical Assumption Test Results:

1. Assumption: "Payment gateway API always returns within 2 seconds"
   Location: src/payment/gateway.js:45-78
   Source: Initial implementation, never validated
   Risk: Critical (payment failures impact revenue)
   Test Strategy: Load testing + network simulation
   Result: ❌ ASSUMPTION FALSE
   - Gateway responses vary from 200ms to 8 seconds
   - 5% of requests exceed 2-second threshold
   - Timeouts occur during peak business hours
   Impact: Payment timeouts cause failed transactions
   Recommendation: Implement timeout handling and retry logic

2. Assumption: "Currency conversion rates are always available"
   Location: src/payment/currency.js:112-145
   Source: Third-party service documentation
   Risk: High (international payments fail)
   Test Strategy: Service failure simulation
   Result: ⚠️ PARTIALLY VALID
   - Rates available 99.5% of time
   - Service outages occur approximately 4 hours/month
   - No fallback mechanism implemented
   Impact: International payments fail during outages
   Recommendation: Add caching and fallback rate sources

3. Assumption: "Users always complete payment in single session"
   Location: src/payment/checkout.js:89-134
   Source: Initial user research (6 months old)
   Risk: Medium (abandoned carts, user frustration)
   Test Strategy: Analytics review + user testing
   Result: ❌ ASSUMPTION FALSE
   - 35% of users abandon and return later
   - Session timeout causes payment data loss
   - No save/resume functionality
   Impact: User frustration and lost revenue
   Recommendation: Implement payment session persistence

4. Assumption: "Fraud detection rules catch all fraudulent transactions"
   Location: src/payment/fraud.js:56-89
   Source: Third-party fraud service claims
   Risk: Critical (financial loss)
   Test Strategy: Historical data analysis + test transactions
   Result: ⚠️ PARTIALLY VALID
   - Catches 92% of known fraud patterns
   - Misses new fraud techniques
   - False positive rate: 1.2%
   Impact: Some fraud gets through, legitimate transactions declined
   Recommendation: Regular fraud rule review and updating

5. Assumption: "Database transactions always rollback on failure"
   Location: src/payment/processor.js:203-245
   Source: ORM documentation
   Risk: High (data inconsistency)
   Test Strategy: Failure injection testing
   Result: ✅ ASSUMPTION VALID
   - Transactions properly rollback on all tested failures
   - Database consistency maintained
   - Edge cases handled correctly
   Impact: None (assumption correct)
   Recommendation: Continue current implementation

Assumption Risk Summary:
  - Critical Assumptions: 8 total, 2 false, 3 partially valid, 3 valid
  - High Risk Assumptions: 12 total, 3 false, 4 partially valid, 5 valid
  - Medium Risk Assumptions: 6 total, 1 false, 2 partially valid, 3 valid
  - Low Risk Assumptions: 2 total, 0 false, 0 partially valid, 2 valid

False Assumption Impact Analysis:
  - Immediate Risks: Payment timeouts causing failed transactions
  - Short-term Risks: International payment failures during outages
  - Long-term Risks: Fraud gaps and user frustration
  - Financial Impact: Estimated $45,000/month in lost/recovered revenue
  - Reputation Impact: User trust erosion due to payment issues

Validation Coverage:
  - Technically testable assumptions: 85% validated
  - Business logic assumptions: 60% validated
  - User experience assumptions: 40% validated
  - Operational assumptions: 70% validated

Recommended Actions:
  1. High Priority: Fix payment timeout handling (critical risk)
  2. High Priority: Add currency rate fallback mechanism
  3. Medium Priority: Implement payment session persistence
  4. Medium Priority: Enhance fraud detection rule updates
  5. Low Priority: Document all validated assumptions

Assumption Testing ROI:
  - Testing time: 3 hours
  - Issues prevented: 4 critical assumptions found false
  - Potential cost savings: $540,000/year (estimated)
  - Risk reduction: Significant reduction in payment failures
  - Confidence increase: Much higher understanding of system limits

Notes

• Assumption testing is most valuable when assumptions are implicit and untested
• Start with high-risk, high-impact assumptions
• Document assumptions as you discover them for future reference
• Re-test assumptions periodically as systems and contexts change
• Share assumption testing findings across the team
• Use assumption testing to inform design and architecture decisions
• Balance assumption testing effort with other testing activities
• Consider cultural and domain-specific assumptions
• Assumption testing complements other testing types but focuses on beliefs rather than specifications
• The most dangerous assumptions are those we don't know we're making