Agent Skills: Speed Monitoring Skill

Speed Monitoring Skill

Purpose

Provides patterns for continuous performance monitoring to prevent speed regressions and maintain Core Web Vitals scores. Establishes automated testing, real user monitoring, and alerting systems to catch performance issues before they impact users.

Core Rules

1. Budget before build — Set performance budgets upfront in budget.json
2. Monitor real users — Lab data ≠ field data, track both synthetic and RUM
3. Alert on regression — Catch issues before users do with threshold-based alerts
4. Block bad deploys — CI fails on performance regression via Lighthouse CI
5. Track over time — Trends matter more than snapshots, store historical data
6. Measure Core Web Vitals — LCP, CLS, INP, FCP, TTFB in production
7. Automated testing — Lighthouse CI on every PR and main branch push
8. Bundle size limits — Enforce JS/CSS budgets in CI pipeline
9. Real user data — Use web-vitals library + beacon API for field data
10. Dashboard visibility — Make performance metrics accessible to team

Performance Budget Targets

| Metric | Good | Warning | Critical | |--------|------|---------|----------| | LCP | <2.5s | 2.5-4s | >4s | | FCP | <1.8s | 1.8-3s | >3s | | CLS | <0.1 | 0.1-0.25 | >0.25 | | INP | <200ms | 200-500ms | >500ms | | TBT | <200ms | 200-600ms | >600ms |

Resource Budgets

| Resource Type | Budget (gzip) | |--------------|---------------| | JavaScript | 200KB | | CSS | 50KB | | Images | 500KB | | Fonts | 100KB | | Total | 800KB |

Monitoring Stack

Synthetic (Lab Data)

• Lighthouse CI in GitHub Actions
• Scheduled cron jobs for production monitoring
• Budget assertions in CI/CD pipeline

Real User (Field Data)

• web-vitals library for Core Web Vitals
• Navigation Timing API for page metrics
• Resource Timing API for asset analysis
• sendBeacon for reliable data transmission

Alerting Channels

• Slack webhooks for warnings
• PagerDuty for critical performance issues
• Email digests for weekly reports

Implementation Checklist

Initial Setup:

• [ ] Create budget.json with performance targets
• [ ] Add lighthouserc.json configuration
• [ ] Set up Lighthouse CI GitHub Action
• [ ] Configure performance assertions (min score 90)

Real User Monitoring:

• [ ] Install web-vitals package
• [ ] Implement vitals tracking client-side
• [ ] Create /api/vitals endpoint
• [ ] Set up database for metrics storage

Bundle Monitoring:

• [ ] Create bundle-stats.js script
• [ ] Add bundle size check to CI
• [ ] Set gzip size budgets

Alerting:

• [ ] Configure Slack webhook
• [ ] Set performance thresholds
• [ ] Implement alert logic in API
• [ ] Test alert notifications

Dashboard:

• [ ] Create performance dashboard page
• [ ] Display current Core Web Vitals
• [ ] Show historical trends
• [ ] Add budget status indicators

References

• Performance Budgets - Budget configuration and bundle monitoring
• Lighthouse CI - CI setup, workflow, and configuration
• Web Vitals Monitoring - RUM implementation and tracking
• Performance Alerts - Threshold configuration and alerting setup
• Synthetic Monitoring - Cron jobs and dashboards

Forbidden

• No performance monitoring in production
• Ignoring field data (RUM)
• No CI performance checks
• Undefined performance budgets
• Manual-only Lighthouse runs
• Silencing performance alerts

Definition of Done

• [ ] Performance budgets defined and documented
• [ ] Lighthouse CI running in GitHub Actions
• [ ] CI fails on performance regression (score <90)
• [ ] Web Vitals tracked in production
• [ ] Bundle size monitoring in CI
• [ ] Alerts configured for poor scores
• [ ] Weekly performance review scheduled
• [ ] Performance dashboard accessible to team