majiayu000 Stress Testing Skill
TASK-338 | Priority: P0 Critical
Rigorously test FlowState to find issues that other testing and agents missed. Covers reliability, backup systems, container stability, security, and data integrity.
Quick Start
# Run all stress tests
npm run test:stress
# Quick smoke test (5 min)
npm run test:stress -- --grep "@quick"
# Full test suite
npm run test:stress -- --timeout=120000
# By category
npm run test:stress -- --grep "Security"
npm run test:stress -- --grep "Data Integrity"
npm run test:stress -- --grep "Restore"
Pre-Flight Checks
Before running stress tests, verify prerequisites:
# 1. Dev server running
curl -s http://localhost:5546 > /dev/null && echo "OK: Dev server" || echo "FAIL: Run npm run dev"
# 2. Docker containers healthy
docker ps --filter "name=supabase" --format "{{.Names}}: {{.Status}}" 2>/dev/null | head -3
# 3. Recent backup exists
ls -la public/shadow-latest.json 2>/dev/null || echo "WARN: No shadow backup"
Test Categories
1. Data Integrity Tests (data-integrity.spec.ts)
| Test | Description | Command |
|---|---|---|
| Rapid Task Creation | Create 20 tasks rapidly, verify no duplicates | --grep "Rapid Creation" |
| Canvas Position | Drag, refresh, verify position persists | --grep "Position Persistence" |
| Concurrent Edits | Edit same task in 2 tabs simultaneously | --grep "Concurrent Edit" |
| Network Instability | Create tasks offline, verify sync on reconnect | --grep "Network Instability" |
2. Security Tests (security.spec.ts)
| Test | Description | Command |
|---|---|---|
| XSS Script Tag | Inject <script> in task title |
--grep "Script Tag" |
| XSS Event Handler | Inject onerror, onclick handlers |
--grep "Event Handler" |
| XSS Rich Text | Test TipTap editor for XSS | --grep "Rich Text Editor" |
| Input Length | 100KB input stress test | --grep "Input Length" |
| SQL Injection | Test search with SQL payloads | --grep "SQL Injection" |
3. Restore Verification Tests (restore-verification.spec.ts)
| Test | Description | Command |
|---|---|---|
| Backup Accessible | Verify backup system is reachable | --grep "accessible" |
| Golden Backup | Verify golden backup has valid structure | --grep "Golden backup" |
| Checksum Valid | Verify backup checksum integrity | --grep "Checksum" |
| Full Restore Cycle | Create -> backup -> delete -> restore -> verify | --grep "Full restore" |
| Shadow Backup | Verify shadow-latest.json is valid | --grep "Shadow backup" |
4. Container Stability Tests (container-stability.spec.ts)
| Test | Description | Command |
|---|---|---|
| Docker Health | Verify Supabase containers healthy | --grep "Docker" |
| DB Restart | Kill DB container, verify app recovers | --grep "DB Restart" |
| Full Stack Restart | Restart entire stack, verify no data loss | --grep "Full Stack" |
| App Reconnection | Verify app auto-reconnects after container restart | --grep "Reconnection" |
5. Performance Benchmarks (store-operations.bench.ts)
| Benchmark | Target | Command |
|---|---|---|
| Create 100 tasks | < 100ms | npm run test:bench |
| Filter 10k tasks | < 50ms | |
| Sort 1k tasks | < 20ms | |
| Find by ID (10k) | < 5ms | |
| JSON serialize 1k | < 50ms | |
| Canvas viewport filter | < 10ms |
Test Matrix: Completed TASKs Coverage
This matrix maps completed tasks to their stress test coverage:
| TASK | Description | Test Coverage |
|---|---|---|
| TASK-131 | Position reset during session | data-integrity.spec.ts: Position Persistence |
| TASK-142 | Position reset on refresh | data-integrity.spec.ts: Position Persistence |
| TASK-255 | Canvas geometry invariants | geometry-invariants.test.ts |
| TASK-256 | Canvas geometry tests | geometry-invariants.test.ts |
| TASK-309-B | Undo/redo system | qa-testing: Memory Leak Check |
| TASK-334 | Completion protocol | Enforced via hooks |
| TASK-335 | Canvas distribution | data-integrity.spec.ts |
| TASK-365 | Restore verification | restore-verification.spec.ts |
Running the Full Suite
# 1. Ensure dev server is running
npm run dev &
# 2. Run all stress tests
npm run test:stress
# 3. Run benchmarks
npm run test:bench
# 4. Generate report
npm run test:stress:report
Report Generation
After running tests, a report is generated at reports/stress-test-report/index.html.
Report contents:
- Test pass/fail counts by category
- Timing metrics
- Screenshots of failures
- Console error captures
- Recommendations
Integration with QA Testing
This skill extends qa-testing with:
- Deeper coverage: Multi-tab scenarios, network partitions
- Container testing: Docker/Supabase health checks
- Security focus: XSS, SQL injection, input fuzzing
- Performance baselines: Benchmarks with thresholds
Chaining:
stress-tester (finds issues) -> dev-debugging (fixes) -> qa-testing (verifies fix)
Failure Response Protocol
Critical Failure (Tier 1)
- Data loss, security vulnerability, container crash
- Action: STOP release, fix immediately
High Failure (Tier 2)
- Performance regression, UI glitches under stress
- Action: Schedule fix within sprint
Medium Failure (Tier 3)
- Edge case failures, minor memory growth
- Action: Add to backlog
Manual Testing Checklist
For issues that can't be automated:
Container Resilience
-
docker kill supabase_db_flow-state-> app shows error, auto-reconnects when container restarts -
npx supabase stop && npx supabase start-> no data loss
Multi-Device Sync
- Open app in Chrome and Firefox simultaneously
- Create task in Chrome -> appears in Firefox within 5s
- Edit same task in both -> last-write-wins, no duplicates
Long Session Stability
- Leave app running 4 hours with periodic activity
- Memory growth < 100MB
- No console errors accumulating
Files
tests/stress/
├── playwright.stress.config.ts # Playwright config for stress tests
├── data-integrity.spec.ts # CRUD, position, concurrent edits
├── security.spec.ts # XSS, SQL injection, input fuzzing
├── restore-verification.spec.ts # Backup/restore cycle tests
├── container-stability.spec.ts # Docker/Supabase health tests
└── store-operations.bench.ts # Performance benchmarks
scripts/
├── run-stress-tests.sh # Full test runner
└── generate-stress-report.cjs # Report generator
Version History
- v1.0 (Jan 2026): Initial release with full test suite