Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when reviewing content pages for user experience and SEO. Applies to blog posts, landing pages, how-to guides, and any long-form content intended for a broad audience.
.claude/skills/thedaviddias-reading-level/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-12 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-17 | ✗→✓ | ▲ Improved | 6% | 0% |
| case-18 | ✗→✓ | ▲ Improved | -62% | 0% |
| case-14 | ✓→✓ | ＝ Same ✓ | -35% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -29% | 0% |
Content written beyond the reading level of the target audience increases bounce rates and reduces time-on-page—both engagement signals that correlate with search rankings.
Analyse this page's content for readability. Calculate or estimate the Flesch-Kincaid Grade Level. Identify sentences longer than 25 words, paragraphs longer than 4 sentences, instances of passive voice, and domain jargon that could be simplified. Report the estimated reading grade level.
Simplify the identified sentences and paragraphs: break long sentences at conjunctions, replace passive voice with active voice, substitute jargon with plain-language equivalents. Target a Flesch Reading Ease score of 60+ (Grade 8 or below for general audiences). Use subheadings to break up text every 200–300 words.
Readability measures how easy content is to understand. When content is written above the typical reading level of the audience, users leave faster (higher bounce rate) and spend less time reading (lower dwell time). These engagement signals influence search rankings indirectly. Clear writing also improves conversion rates and user satisfaction.
Extract the text content from the page body (excluding navigation, headers, footers). Calculate or estimate the Flesch-Kincaid Grade Level. Count: sentences longer than 25 words, paragraphs with more than 4 sentences, passive voice instances, and words with more than 3 syllables that could be simplified. Report the grade level and top 5 most complex sentences.
For full implementation details, code examples, and framework-specific guidance, see references/rule.md.
Rule page: https://frontendchecklist.io/en/rules/seo/reading-level
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-14 | pass→pass | 6,716 | 1,959 | -71% | 1 | 1 | 0% | 1,187 | 777 | -35% | 0 | 0 | — |
case-01 | pass→pass | 9,878 | 3,820 | -61% | 1 | 1 | 0% | 1,692 | 1,206 | -29% | 0 | 0 | — |
case-02 | pass→pass | 8,372 | 3,316 | -60% | 1 | 1 | 0% | 1,513 | 1,043 | -31% | 0 | 0 | — |
case-03 | pass→pass | 8,392 | 3,197 | -62% | 1 | 1 | 0% | 1,461 | 1,057 | -28% | 0 | 0 | — |
case-04 | pass→pass | 13,270 | 4,876 | -63% | 1 | 1 | 0% | 2,337 | 1,406 | -40% | 0 | 0 | — |
case-05 | pass→pass | 7,064 | 3,477 | -51% | 1 | 1 | 0% | 1,265 | 1,030 | -19% | 0 | 0 | — |
case-06 | pass→pass | 7,847 | 6,749 | -14% | 1 | 1 | 0% | 1,470 | 1,804 | +23% | 0 | 0 | — |
case-07 | pass→pass | 4,368 | 3,584 | -18% | 1 | 1 | 0% | 879 | 1,144 | +30% | 0 | 0 | — |
case-08 | pass→pass | 7,927 | 4,425 | -44% | 1 | 1 | 0% | 1,447 | 1,314 | -9% | 0 | 0 | — |
case-09 | pass→pass | 11,051 | 7,727 | -30% | 1 | 1 | 0% | 2,054 | 1,839 | -10% | 0 | 0 | — |
case-10 | pass→pass | 13,567 | 8,375 | -38% | 1 | 1 | 0% | 2,230 | 1,984 | -11% | 0 | 0 | — |
case-11 | fail→fail | 9,221 | 1,753 | -81% | 1 | 1 | 0% | 1,533 | 735 | -52% | 0 | 0 | — |
case-12 | fail→pass | 14,476 | 11,015 | -24% | 1 | 1 | 0% | 2,281 | 2,262 | -1% | 0 | 0 | — |
case-13 | pass→pass | 13,242 | 9,553 | -28% | 1 | 1 | 0% | 2,090 | 2,153 | +3% | 0 | 0 | — |
case-15 | pass→pass | 6,359 | 2,333 | -63% | 1 | 1 | 0% | 1,129 | 716 | -37% | 0 | 0 | — |
case-16 | pass→pass | 5,259 | 5,037 | -4% | 1 | 1 | 0% | 930 | 1,299 | +40% | 0 | 0 | — |
case-17 | fail→pass | 5,931 | 3,755 | -37% | 1 | 1 | 0% | 1,047 | 1,115 | +6% | 0 | 0 | — |
case-18 | fail→pass | 15,879 | 3,139 | -80% | 1 | 1 | 0% | 2,942 | 1,117 | -62% | 0 | 0 | — |
case-19 | pass→pass | 9,407 | 1,911 | -80% | 1 | 1 | 0% | 1,601 | 818 | -49% | 0 | 0 | — |
case-20 | pass→pass | 12,824 | 7,924 | -38% | 1 | 1 | 0% | 2,083 | 1,702 | -18% | 0 | 0 | — |
case-21 | pass→pass | 7,663 | 4,121 | -46% | 1 | 1 | 0% | 1,293 | 1,070 | -17% | 0 | 0 | — |
case-22 | fail→fail | 12,611 | 8,222 | -35% | 1 | 1 | 0% | 2,046 | 1,832 | -10% | 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. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.