data.cms.gov: Lighthouse — week 2026-W38
Trend
| Week | Performance | Accessibility | Best practices | SEO |
|---|---|---|---|---|
| 2026-W24 | 46 | 96 | 100 | 92 |
| 2026-W25 | 49 | 95 | 100 | 92 |
| 2026-W26 | 47 | 95 | 100 | 92 |
| 2026-W27 | 47 | 95 | 100 | 92 |
| 2026-W28 | 47 | 95 | 100 | 92 |
| 2026-W29 | 47 | 95 | 100 | 92 |
| 2026-W30 | 45 | 95 | 100 | 92 |
| 2026-W31 | 45 | 95 | 100 | 92 |
| 2026-W32 | 47 | 95 | 100 | 92 |
| 2026-W33 | 45 | 95 | 100 | 92 |
| 2026-W34 | 49 | 95 | 100 | 92 |
| 2026-W35 | 48 | 95 | 100 | 92 |
| 2026-W36 | 48 | 95 | 100 | 92 |
| 2026-W37 | 49 | 95 | 100 | 92 |
| 2026-W38 | 49 | 95 | 100 | 92 |
| 2026-W39 | 48 | 95 | 100 | 92 |
| 2026-W40 | 49 | 98 | 100 | 92 |
Performance impact
- Avg extra LCP over 2.5s
- 10.2s
- Avg extra weight over 1.6 MB
- 3.15 MB
Medians across sampled pages
- Performance
- 49/100
- Accessibility
- 95/100
- Best practices
- 100/100
- SEO
- 92/100
- Agentic browsing
- 50/100
- Largest Contentful Paint
- 10.6s
- First Contentful Paint
- 3.4s
- Speed Index
- 4.2s
- Total Blocking Time
- 0.7s
- Cumulative Layout Shift
- 0.001
Recommendations
Performance
| Recommendation | Pages | Est. saving |
|---|---|---|
| Reduce unused JavaScript | 247/269 | 225.0 MB · 1034.5s |
| Render-blocking requests | 247/269 | 306.6s |
| Reduce unused CSS | 247/269 | 25.3 MB · 136.3s |
| First Contentful Paint | 247/269 | — |
| Largest Contentful Paint | 247/269 | — |
| Time to Interactive | 247/269 | — |
| Page prevented back/forward cache restoration | 247/269 | — |
| Network dependency tree | 247/269 | — |
| Minimizes main-thread work | 246/269 | 187.2s |
| JavaScript execution time | 246/269 | 173.1s |
| Use efficient cache lifetimes | 229/269 | 62.6s |
| Speed Index | 226/269 | — |
| Total Blocking Time | 222/269 | — |
| Max Potential First Input Delay | 219/269 | — |
| Font display | 159/269 | 14.3s |
| Legacy JavaScript | 91/269 | 13.7s |
| Layout shift culprits | 80/269 | — |
| Forced reflow | 60/269 | — |
| LCP request discovery | 45/269 | — |
| Reduce initial server response time | 28/269 | 3.2s |
| Improve image delivery | 19/269 | 2.9s |
| LCP breakdown | 4/269 | — |
| Document request latency | 3/269 | 1.7s |
| Avoid multiple page redirects | 2/269 | 1.9s |
| Duplicated JavaScript | 1/269 | 0.1s |
Best Practices
| Recommendation | Pages | Est. saving |
|---|---|---|
| Browser errors were logged to the console | 98/269 | — |
| Missing source maps for large first-party JavaScript | 93/269 | — |
| Uses third-party cookies | 21/269 | — |
| Issues were logged in the `Issues` panel in Chrome Devtools | 21/269 | — |
SEO
| Recommendation | Pages | Est. saving |
|---|---|---|
| Document does not have a meta description | 179/269 | — |
| Links are not crawlable | 155/269 | — |
| Page is blocked from indexing | 81/269 | — |
| Links do not have descriptive text | 5/269 | — |
Agentic (AI-readiness)
| Recommendation | Pages | Est. saving |
|---|---|---|
| WebMCP tools registered | 247/269 | — |
| WebMCP form coverage | 247/269 | — |
| WebMCP schemas are valid | 247/269 | — |
| llms.txt follows recommendations | 247/269 | — |
| Accessibility tree is not well-formed | 167/269 | — |
| Cumulative Layout Shift | 80/269 | — |
What the Agentic score means
The Agentic (AI-readiness) score is new in Google Lighthouse (13.4+). It measures how well a page works for AI agents and assistants — the tools that increasingly mediate how people find and use government services. It checks for things like an llms.txt file (a machine-readable guide for language models), valid structured data, a well-formed accessibility tree that agents can parse, and WebMCP tool/form descriptions. A higher score means an AI assistant is more likely to understand the page and complete tasks on a citizen's behalf correctly.