Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Professional-grade brand voice analysis, SEO optimization, and platform-specific content frameworks.
.claude/skills/sickn33-content-creator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 2379% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 51% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 60% | 0% |
Draft and review audience-specific content using supplied brand examples, local text diagnostics, and adaptable channel templates.
Use this skill when writing blog posts, creating social media content, establishing brand voice, optimizing content for SEO, or planning content calendars.
content creation, blog posts, SEO, brand voice, social media, content calendar, marketing content, content strategy, content marketing, brand consistency, content optimization, social media marketing, content planning, blog writing, content frameworks, brand guidelines, social media strategy
Obtain the audience, purpose, approved claims and sources, brand examples, channel, and desired next action. Reuse supplied constraints; do not invent audience research. Python 3 is sufficient for the optional local scripts. Run the examples from this skill directory with a permitted UTF-8 input below the working directory (maximum 1 MiB); paths elsewhere are rejected. The scripts read that file and print diagnostics, without calling an analytics service or editing it. Review any private text before sharing the output. Drafting does not authorize scheduling, sending or publication.
scripts/brand_voice_analyzer.py on existing content to record rough lexical featuresreferences/brand_guidelines.md to select voice attributesreferences/content_frameworks.mdscripts/seo_optimizer.py [file] [primary-keyword] to optimizereferences/social_media_optimization.mdreferences/content_frameworks.mdassets/content_calendar_template.mdWhen creating content for a new brand or client:
bash python scripts/brand_voice_analyzer.py existing_content.txt
references/brand_guidelines.mdreferences/content_frameworks.mdbash python scripts/seo_optimizer.py blog_post.md "primary keyword" "secondary,keywords,list"
references/social_media_optimization.mdreferences/content_frameworks.mdassets/content_calendar_template.mdCounts a small English vocabulary and estimates sentence length/readability. It cannot establish authentic brand voice or validate factual claims.
Usage: python scripts/brand_voice_analyzer.py <file> [json|text]
Returns:
Analyzes content for SEO optimization and provides actionable recommendations.
Usage: python scripts/seo_optimizer.py <file> [primary_keyword] [secondary_keywords]
Returns:
references/brand_guidelines.md
references/content_frameworks.md
references/social_media_optimization.md
Track these KPIs for content success:
This skill works best with:
bash# Analyze brand voice python scripts/brand_voice_analyzer.py content.txt # Optimize for SEO python scripts/seo_optimizer.py article.md "main keyword" # Compare the draft manually with approved brand examples and prohibited claims # Create monthly calendar cp assets/content_calendar_template.md this_month_calendar.md
For an approved release note stating “CSV export is now available”, prepare a short help article explaining where export lives and a social draft linking to that article. Keep the exact supported formats and limitations from the source. Do not turn the claim into “save hours” without measured evidence. Run the local text diagnostics, then verify the instructions against the actual product and compare tone with two approved posts. Return both drafts, source links, unresolved facts and a proposed calendar slot. Expected result: reviewable copy; no posts have been sent.
ties or absent matches do not identify a brand personality. It is not a validated reading assessment and does not support multilingual scoring reliably.
search demand, accessibility, ranking, or causality. Character lengths and word counts are observations, not quality thresholds.
client facts. Confirm rights and approvals for quoted material and images.
the selected format; use account analytics to test timing instead of universal rules.
Google's people-first content guidance explains why satisfying reader needs matters more than filling a target word count.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 6,083 | 12,764 | +110% | 1 | 1 | 0% | 336 | 4,001 | +1091% | 0 | 0 | — |
case-02 | fail→fail | 4,520 | 9,067 | +101% | 1 | 1 | 0% | 804 | 3,401 | +323% | 0 | 0 | — |
case-03 | fail→pass | 3,095 | 16,755 | +441% | 1 | 1 | 0% | 203 | 5,033 | +2379% | 0 | 0 | — |
case-04 | fail→pass | 6,623 | 2,105 | -68% | 1 | 1 | 0% | 1,208 | 2,030 | +68% | 0 | 0 | — |
case-05 | fail→pass | 7,704 | 2,817 | -63% | 1 | 1 | 0% | 1,417 | 2,141 | +51% | 0 | 0 | — |
case-06 | fail→pass | 10,393 | 3,260 | -69% | 1 | 1 | 0% | 1,829 | 2,215 | +21% | 0 | 0 | — |
case-07 | fail→pass | 7,562 | 3,330 | -56% | 1 | 1 | 0% | 1,395 | 2,234 | +60% | 0 | 0 | — |
case-08 | pass→pass | 9,435 | 2,426 | -74% | 1 | 1 | 0% | 1,626 | 2,099 | +29% | 0 | 0 | — |
case-09 | fail→pass | 4,986 | 2,545 | -49% | 1 | 1 | 0% | 804 | 2,090 | +160% | 0 | 0 | — |
case-10 | pass→pass | 9,160 | 3,317 | -64% | 1 | 1 | 0% | 1,811 | 2,213 | +22% | 0 | 0 | — |
case-11 | pass→pass | 9,817 | 1,953 | -80% | 1 | 1 | 0% | 1,539 | 1,999 | +30% | 0 | 0 | — |
case-12 | fail→pass | 6,122 | 1,777 | -71% | 1 | 1 | 0% | 1,245 | 2,036 | +64% | 0 | 0 | — |
case-13 | fail→pass | 7,608 | 2,890 | -62% | 1 | 1 | 0% | 1,226 | 2,185 | +78% | 0 | 0 | — |
case-14 | fail→pass | 3,018 | 1,946 | -36% | 1 | 1 | 0% | 517 | 2,056 | +298% | 0 | 0 | — |
case-15 | fail→pass | 8,633 | 2,450 | -72% | 1 | 1 | 0% | 1,628 | 2,108 | +29% | 0 | 0 | — |
case-16 | fail→pass | 4,482 | 1,548 | -65% | 1 | 1 | 0% | 822 | 1,954 | +138% | 0 | 0 | — |
case-17 | pass→pass | 3,823 | 3,347 | -12% | 1 | 1 | 0% | 598 | 2,224 | +272% | 0 | 0 | — |
case-18 | fail→pass | 6,387 | 2,043 | -68% | 1 | 1 | 0% | 943 | 1,978 | +110% | 0 | 0 | — |
case-19 | fail→fail | 11,370 | 6,654 | -41% | 1 | 1 | 0% | 1,843 | 2,715 | +47% | 0 | 0 | — |
case-20 | fail→pass | 13,776 | 4,247 | -69% | 1 | 1 | 0% | 2,575 | 2,347 | -9% | 0 | 0 | — |
case-21 | fail→pass | 15,900 | 5,486 | -65% | 1 | 1 | 0% | 3,299 | 2,560 | -22% | 0 | 0 | — |
case-22 | fail→fail | 30,105 | 16,988 | -44% | 1 | 1 | 0% | 1,924 | 5,585 | +190% | 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, and 20 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +64 percentage points is the difference between those two pass rates over the 20 comparable cases.
The publisher has shipped newer versions since this run, so these numbers describe v1, not the version currently listed.
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.