farmountain Skill: Performance Tester
Purpose
Execute performance tests to validate application scalability, response times, and resource utilization under various load conditions.
Core Capabilities
- Load Testing: Simulate user traffic at scale
- Stress Testing: Push systems beyond normal operating conditions
- Scalability Validation: Verify performance as load increases
- Resource Monitoring: Track CPU, memory, network, and disk usage
- Bottleneck Identification: Pinpoint performance constraints
Position Card Protocol
SPEC Card (Requirements)
Performance Tester - Performance Requirements
├── Load Profiles: [user count, request patterns, duration]
├── Performance Targets: [response time, throughput, error rate]
├── System Resources: [CPU, memory, network, storage limits]
├── Test Scenarios: [peak load, sustained load, spike testing]
├── Monitoring Scope: [application, infrastructure, dependencies]
├── Success Criteria: [targets met under specified conditions]
└── Failure Thresholds: [when to stop testing]
TEST Card (Validation)
Performance Tester - Validation Checks
├── Load Generation: [can simulate required user load]
├── Monitoring Setup: [metrics collection configured]
├── Resource Limits: [test environment capacity validated]
├── Scenario Definition: [load patterns properly specified]
├── Result Collection: [metrics captured accurately]
├── Environment Isolation: [no impact on other systems]
└── Safety Controls: [automatic test termination]
SOLVER Card (Implementation)
Performance Tester - Execution Strategy
├── Load Modeling: [understand target user behavior]
├── Test Planning: [sequence load scenarios logically]
├── Environment Scaling: [ensure adequate test resources]
├── Monitoring Deployment: [setup comprehensive metrics]
├── Scenario Execution: [run performance test scenarios]
├── Result Analysis: [identify bottlenecks and limits]
└── Report Generation: [create actionable insights]
SKEPTIC Card (Risk Assessment)
Performance Tester - Risk Analysis
├── Resource Exhaustion: [test environment overwhelmed]
├── Production Impact: [testing affects live systems]
├── Cost Overruns: [expensive test infrastructure]
├── False Results: [inaccurate load simulation]
├── Environment Differences: [test vs production gaps]
├── Test Interference: [performance tests affect each other]
└── Maintenance Complexity: [complex test script upkeep]
VERIFIER Card (Evidence Requirements)
Performance Tester - Verification Evidence
├── Load Achievement: [required user load successfully generated]
├── Target Validation: [performance targets met or exceeded]
├── Resource Monitoring: [system resources tracked]
├── Bottleneck Analysis: [performance constraints identified]
├── Scalability Proof: [performance under increasing load]
├── Stability Evidence: [system remained stable under load]
└── Recommendations: [actionable performance improvements]
Workflow Integration
Invoked By
- NFR Agent: When validating non-functional requirements
- Release Manager: Before performance-sensitive releases
- DevOps Platform: During infrastructure scaling decisions
- Cost Estimator: When evaluating cloud resource needs
- Build Validator: As part of comprehensive quality gates
Invokes
- Observability Agent: To enhance monitoring during testing
- Metrics Agent: To track performance metrics over time
- Drift Detector: To identify performance regression
- Confidence Agent: To calibrate confidence in performance claims
Evidence-Gated Outputs
Primary Output: Performance Test Report
# Performance Test Execution Report
## Test Configuration
- **Test Type**: [load/stress/scalability/volume/spike]
- **Duration**: [total test time]
- **Load Profile**: [user count, ramp-up pattern, sustained period]
- **Target System**: [application version, environment details]
## Performance Results
### Response Time Metrics
- **Average Response Time**: [milliseconds]
- **95th Percentile**: [milliseconds] (Target: [threshold])
- **99th Percentile**: [milliseconds] (Target: [threshold])
- **Maximum Response Time**: [milliseconds]
### Throughput Metrics
- **Requests per Second**: [count] (Target: [threshold])
- **Successful Requests**: [percentage]% (Target: [threshold]%)
- **Error Rate**: [percentage]% (Target: < [threshold]%)
### Resource Utilization
- **CPU Usage**: Average [percentage]%, Peak [percentage]%
- **Memory Usage**: Average [percentage]%, Peak [percentage]%
- **Network I/O**: [Mbps] sent, [Mbps] received
- **Disk I/O**: [IOPS] read, [IOPS] write
## Scalability Analysis
### Load Progression Results
| Concurrent Users | Response Time (P95) | Throughput (RPS) | Error Rate | CPU % | Memory % |
|------------------|---------------------|------------------|------------|-------|----------|
| [count] | [time] | [count] | [rate] | [pct] | [pct] |
### Bottleneck Identification
- **Primary Bottleneck**: [component limiting performance]
- **Secondary Bottlenecks**: [additional constraints]
- **Resource Limits Hit**: [CPU/memory/network/disk thresholds]
## Failure Analysis
### Performance Degradation Points
- [Load levels where performance dropped significantly]
- [Components that failed under load]
- [Error patterns observed]
### Root Cause Analysis
- [Technical reasons for performance issues]
- [Infrastructure limitations identified]
- [Application bottlenecks found]
## Recommendations
### Immediate Actions
- [Critical performance fixes needed]
- [Infrastructure scaling requirements]
- [Configuration optimizations]
### Long-term Improvements
- [Architecture changes recommended]
- [Code optimization opportunities]
- [Monitoring enhancements needed]
Secondary Outputs
- Performance Graphs: Visual charts of metrics over time
- Load Profiles: Detailed load generation patterns
- Resource Monitoring Logs: Comprehensive system metrics
- Bottleneck Analysis Reports: Detailed technical analysis
Failure Modes & Recovery
FM-001: Load Generation Failure
Trigger: Cannot achieve required user load or request rates Recovery: Scale test infrastructure, optimize load generation, reduce load requirements Fallback: Test at maximum achievable load with reduced evidence quality
FM-002: Resource Exhaustion
Trigger: Test environment runs out of CPU, memory, or other resources Recovery: Increase test infrastructure capacity, optimize test scenarios, implement resource monitoring Fallback: Reduce test scope, monitor available resources only
FM-003: Monitoring Gaps
Trigger: Cannot collect required performance metrics Recovery: Enhance monitoring setup, add missing metric collection, validate monitoring tools Fallback: Proceed with available metrics, note monitoring limitations
FM-004: Test Environment Differences
Trigger: Test results don't reflect production performance Recovery: Align test environment with production, document differences, adjust targets accordingly Fallback: Qualify results with environment differences noted
FM-005: Unstable Test Conditions
Trigger: External factors affect test results (network issues, competing load) Recovery: Isolate test environment, implement result validation, run multiple test iterations Fallback: Document test conditions, provide confidence intervals
Quality Gates
Pre-Execution Gates
- ✅ Load profiles defined and achievable
- ✅ Performance targets specified and measurable
- ✅ Test environment provisioned and validated
- ✅ Monitoring and metrics collection configured
- ✅ Safety controls and automatic termination implemented
Post-Execution Gates
- ✅ Required load levels achieved and sustained
- ✅ Performance targets evaluated against results
- ✅ Resource utilization monitored and analyzed
- ✅ Bottlenecks identified and documented
- ✅ Results reproducible and consistent
Evidence Quality Gates
- ✅ Performance metrics captured comprehensively
- ✅ Load conditions documented accurately
- ✅ Analysis performed with technical depth
- ✅ Recommendations actionable and prioritized
Invariants Validated
INV-036: Performance Requirements
Validation: Application performance meets defined targets Evidence: Performance test results against requirements Failure Impact: Poor user experience and scalability issues
INV-037: Resource Efficiency
Validation: Application uses resources efficiently under load Evidence: Resource utilization metrics within acceptable ranges Failure Impact: Excessive infrastructure costs or poor scalability
INV-038: Scalability Validation
Validation: Application scales appropriately with load Evidence: Performance maintained as load increases Failure Impact: Inability to handle user growth
Risk Mitigation
Performance Risks
- Unrealistic Testing: Representative load profiles prevent false confidence
- Environment Gaps: Production-like test environments ensure result validity
- Monitoring Blind Spots: Comprehensive metrics prevent undetected issues
Operational Risks
- Resource Overload: Safety controls prevent test environment damage
- Cost Control: Efficient test execution prevents budget overruns
- Result Accuracy: Multiple test runs and statistical analysis improve confidence
Business Risks
- Performance Regression: Continuous testing prevents unnoticed degradation
- Scalability Limits: Early bottleneck detection enables proactive scaling
- User Impact: Performance validation prevents production issues
Evolution Path
Phase 1 (Current): Load and Performance Testing
- Load generation and monitoring
- Performance target validation
- Bottleneck identification
Phase 2 (Future): Intelligent Performance Testing
- AI-powered load pattern generation
- Predictive bottleneck analysis
- Automated performance optimization
Phase 3 (Future): Continuous Performance
- Real-time performance monitoring
- Automated scaling recommendations
- Predictive performance issue detection
d:\All_Projects\sdlc_agent_swarm.agents\skills\performance-tester\skill.md