Check of ciunkos.com/ai-prompt-testing
Nov 7, 2026, 13:41 · daily
- Performance
- Accessibility
- Best Practices
- SEO
Lighthouse 13.5 (mobile, simulated throttling), median of three runs. Scores 0–100: 90 and up green, 50–89 orange, below 50 red; a dash means the category was not measured in that check.
TBT is the lab stand-in for INP, which needs real interactions.
- Performance
- 100good±0 vs previous±0 vs 7-day median
- LCP
- 940 msgood−23 ms vs previous−23 ms vs 7-day median
- CLS
- 0good±0 vs previous±0 vs 7-day median
- TBT
- 0 msgood±0 vs previous±0 vs 7-day median
- FCP
- 940 msgood−23 ms vs previous−23 ms vs 7-day median
- TTFB
- 42 msgood+24 ms vs previous+13 ms vs 7-day median
No slowdown against the previous check and the 7-day median.
- LCP element
section.hero > div.wrap > div > p.leadYou fix one answer in the system prompt and three others quietly get worse. Eva…- Page weight
- 18 KB
- Long tasks
- 1
- Conditions
- Lighthouse 13.5.0, mobile (Moto G Power emulation, 412×823), simulated throttling (150 ms RTT, 1.6 Mbps, CPU 4×), median of 3 runs, benchmarkIndex 1585
Loads
| Load | Perf | LCP | FCP | TBT | CLS | TTFB |
|---|---|---|---|---|---|---|
| 1 | 100 | 991 ms | 991 ms | 0 ms | 0 | 75 ms |
| 2 | 100 | 940 ms | 940 ms | 0 ms | 0 | 42 ms |
| 3 | 100 | 920 ms | 920 ms | 0 ms | 0 | 21 ms |
What failed
Performance 100
- Network dependency tree
network-dependency-tree-insightpreconnect hints help the browser establish a connection earlier in the page loAdd preconnect hints to your most important origins, but try to use no more tha - Render-blocking requests
render-blocking-insighthttps://ciunkos.com/ai-prompt-testing/a/app.css?v=2026-11-03-605276b
Accessibility 100
Nothing failed in this category.
Best Practices 100
Nothing failed in this category.
SEO 100
Nothing failed in this category.
Google PageSpeed Insights: Comparison details
Vitals Desk is the primary measurement. Google adds one mobile lab run from its servers. This comparison does not change our scores, history or slowdown emails.
Substantial differences: LCP, FCP
The Lighthouse versions differ; scores and checks may not be directly comparable.
Different locations, hardware, Lighthouse versions and a single run versus our three-run median can produce different results. These are comparison notes, not regression alerts.
| Metric | Vitals Desk | Google minus Vitals Desk | |
|---|---|---|---|
| Performance | 100 | 99 | −1 |
| Accessibility | 100 | 100 | ±0 |
| Best Practices | 100 | 92 | −8 |
| SEO | 100 | 100 | ±0 |
| LCP | 940 ms | 1.5 s | +591 ms |
| CLS | 0 | 0 | ±0 |
| TBT | 0 ms | 80 ms | +80 ms |
| FCP | 940 ms | 1.5 s | +591 ms |
| TTFB | 42 ms | 1 ms | −41 ms |
- Lighthouse version
- 13.6.0
- Google checked at
- Nov 7, 2026, 13:41
- Google’s final URL
- https://ciunkos.com/ai-prompt-testing
Additional failed checks in Google’s sample
These appear in Google’s saved sample but not in our saved list. Each source keeps a limited list.
- Legacy JavaScript
legacy-javascript-insight(Performance) - Reduce unused JavaScript
unused-javascript(Performance) - Browser errors were logged to the console
errors-in-console(Best Practices) - Issues were logged in the `Issues` panel in Chrome Devtools
inspector-issues(Best Practices)