Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when applies especially to e-commerce checkouts, contact/lead forms, pricing pages, and YMYL (Your Money or Your Life) content. Use when auditing E-E-A-T or conversion optimization.
.claude/skills/thedaviddias-trust-signals/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 148% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -4% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 58% | 0% |
Google's Quality Rater Guidelines explicitly evaluate trustworthiness as a core page quality signal; pages lacking visible trust evidence score lower in manual quality assessments that influence ranking systems.
On key conversion pages (checkout, contact, pricing, landing pages), check for: security/SSL badges near forms, third-party review platform ratings (Google, Trustpilot), industry certifications, client/partner logos, testimonials with names and photos, money-back guarantee language. Verify all badges are genuine (not just images).
Add appropriate trust signals to each key page: embed real review widgets from Google/Trustpilot, display relevant certifications with verifiable links, add named testimonials with photos, include security badges near form submission areas. Ensure all trust elements are current and accurate.
Explain how trust signals affect both user conversion rates and Google's quality assessments, what E-E-A-T means for content quality, and why YMYL pages face higher trust scrutiny from Google's Quality Rater Guidelines.
Review metadata generation, rendered HTML, structured data, and response headers related to Show trust signals on key pages. Flag exact routes or templates where search-facing output violates the rule, and describe how to verify the final page output.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/seo/trust-signals
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-02 | pass→pass | 12,488 | 12,143 | -3% | 1 | 1 | 0% | 2,226 | 2,669 | +20% | 0 | 0 | — |
case-01 | fail→pass | 8,905 | 16,618 | +87% | 1 | 1 | 0% | 1,507 | 3,741 | +148% | 0 | 0 | — |
case-03 | fail→pass | 15,490 | 13,479 | -13% | 1 | 1 | 0% | 2,467 | 2,753 | +12% | 0 | 0 | — |
case-04 | pass→pass | 13,182 | 8,806 | -33% | 1 | 1 | 0% | 2,133 | 1,892 | -11% | 0 | 0 | — |
case-05 | pass→pass | 17,610 | 15,563 | -12% | 1 | 1 | 0% | 3,089 | 3,279 | +6% | 0 | 0 | — |
case-06 | fail→pass | 10,347 | 10,421 | +1% | 1 | 1 | 0% | 1,754 | 2,064 | +18% | 0 | 0 | — |
case-07 | fail→pass | 15,580 | 11,725 | -25% | 1 | 1 | 0% | 2,698 | 2,582 | -4% | 0 | 0 | — |
case-08 | fail→pass | 10,004 | 12,297 | +23% | 1 | 1 | 0% | 1,649 | 2,605 | +58% | 0 | 0 | — |
case-09 | fail→pass | 18,416 | 15,724 | -15% | 1 | 1 | 0% | 2,959 | 3,130 | +6% | 0 | 0 | — |
case-10 | pass→pass | 9,665 | 8,072 | -16% | 1 | 1 | 0% | 1,706 | 1,856 | +9% | 0 | 0 | — |
case-11 | fail→pass | 15,085 | 10,179 | -33% | 1 | 1 | 0% | 2,549 | 2,169 | -15% | 0 | 0 | — |
case-12 | fail→pass | 13,468 | 9,751 | -28% | 1 | 1 | 0% | 2,212 | 1,902 | -14% | 0 | 0 | — |
case-13 | fail→pass | 17,513 | 18,741 | +7% | 1 | 1 | 0% | 2,957 | 4,032 | +36% | 0 | 0 | — |
case-14 | pass→pass | 20,163 | 19,872 | -1% | 1 | 1 | 0% | 3,264 | 3,889 | +19% | 0 | 0 | — |
case-15 | pass→pass | 13,070 | 67,197 | +414% | 1 | 1 | 0% | 2,684 | 1,880 | -30% | 0 | 0 | — |
case-16 | pass→pass | 15,607 | 17,976 | +15% | 1 | 1 | 0% | 2,801 | 3,737 | +33% | 0 | 0 | — |
case-17 | pass→pass | 6,310 | 5,974 | -5% | 1 | 1 | 0% | 1,041 | 1,421 | +37% | 0 | 0 | — |
case-18 | pass→fail | 16,475 | 16,154 | -2% | 1 | 1 | 0% | 2,404 | 3,449 | +43% | 0 | 0 | — |
case-19 | pass→fail | 16,338 | 14,818 | -9% | 1 | 1 | 0% | 2,521 | 2,917 | +16% | 0 | 0 | — |
case-20 | fail→fail | 13,332 | 14,850 | +11% | 1 | 1 | 0% | 2,197 | 3,115 | +42% | 0 | 0 | — |
case-21 | pass→pass | 18,189 | 10,998 | -40% | 1 | 1 | 0% | 3,507 | 2,792 | -20% | 0 | 0 | — |
case-22 | pass→pass | 6,970 | 8,460 | +21% | 1 | 1 | 0% | 1,513 | 2,317 | +53% | 0 | 0 | — |
case-23 | pass→pass | 17,963 | 15,372 | -14% | 1 | 1 | 0% | 3,835 | 3,623 | -6% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +30 percentage points is the difference between those two pass rates over the 23 comparable cases. 2 cases got worse with the skill loaded, and they are included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.