Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Expert social media strategist specializing in cross-platform strategy, professional audience development, and integrated campaign management. Focused on building brand authority across LinkedIn, T...
.claude/skills/dev-dennis-040-marketing-social-media-strategist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 41% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 72% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 26% | 0% |
name: Social Media Strategist description: Expert social media strategist for LinkedIn, Twitter, and professional platforms. Creates cross-platform campaigns, builds communities, manages real-time engagement, and develops thought leadership strategies. tools: WebFetch, WebSearch, Read, Write, Edit color: blue
Expert social media strategist specializing in cross-platform strategy, professional audience development, and integrated campaign management. Focused on building brand authority across LinkedIn, Twitter, and professional social platforms through cohesive messaging, community engagement, and thought leadership.
Use this agent when you need:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 9,618 | 7,790 | -19% | 1 | 1 | 0% | 1,959 | 2,818 | +44% | 0 | 0 | — |
case-02 | pass→pass | 9,996 | 10,784 | +8% | 1 | 1 | 0% | 1,887 | 3,228 | +71% | 0 | 0 | — |
case-03 | fail→fail | 11,403 | 8,826 | -23% | 1 | 1 | 0% | 2,265 | 3,028 | +34% | 0 | 0 | — |
case-04 | fail→fail | 10,879 | 9,279 | -15% | 1 | 1 | 0% | 2,229 | 3,149 | +41% | 0 | 0 | — |
case-05 | fail→pass | 10,087 | 11,034 | +9% | 1 | 1 | 0% | 2,213 | 2,883 | +30% | 0 | 0 | — |
case-06 | fail→fail | 12,593 | 12,236 | -3% | 1 | 1 | 0% | 2,594 | 3,659 | +41% | 0 | 0 | — |
case-07 | pass→pass | 12,450 | 14,489 | +16% | 1 | 1 | 0% | 2,294 | 3,284 | +43% | 0 | 0 | — |
case-08 | pass→pass | 9,985 | 2,549 | -74% | 1 | 1 | 0% | 1,913 | 1,793 | -6% | 0 | 0 | — |
case-09 | fail→pass | 9,016 | 7,746 | -14% | 1 | 1 | 0% | 1,701 | 2,404 | +41% | 0 | 0 | — |
case-10 | pass→pass | 6,945 | 6,614 | -5% | 1 | 1 | 0% | 1,334 | 2,291 | +72% | 0 | 0 | — |
case-11 | fail→pass | 5,472 | 2,234 | -59% | 1 | 1 | 0% | 999 | 1,718 | +72% | 0 | 0 | — |
case-12 | fail→pass | 10,792 | 7,292 | -32% | 1 | 1 | 0% | 2,061 | 2,588 | +26% | 0 | 0 | — |
case-13 | pass→pass | 4,484 | 4,095 | -9% | 1 | 1 | 0% | 833 | 2,100 | +152% | 0 | 0 | — |
case-14 | pass→pass | 6,918 | 7,349 | +6% | 1 | 1 | 0% | 1,364 | 2,618 | +92% | 0 | 0 | — |
case-15 | pass→pass | 13,532 | 15,710 | +16% | 1 | 1 | 0% | 2,362 | 4,225 | +79% | 0 | 0 | — |
case-16 | pass→pass | 15,112 | 9,905 | -34% | 1 | 1 | 0% | 1,979 | 3,074 | +55% | 0 | 0 | — |
case-17 | pass→pass | 14,343 | 15,150 | +6% | 1 | 1 | 0% | 2,543 | 4,102 | +61% | 0 | 0 | — |
case-18 | pass→pass | 12,031 | 10,407 | -13% | 1 | 1 | 0% | 2,131 | 3,173 | +49% | 0 | 0 | — |
case-19 | pass→pass | 14,898 | 14,919 | +0% | 1 | 1 | 0% | 2,768 | 4,021 | +45% | 0 | 0 | — |
case-20 | fail→fail | 6,934 | 4,354 | -37% | 1 | 1 | 0% | 1,365 | 2,169 | +59% | 0 | 0 | — |
case-21 | fail→fail | 17,242 | 16,286 | -6% | 1 | 1 | 0% | 3,760 | 4,547 | +21% | 0 | 0 | — |
case-22 | fail→fail | 8,999 | 13,942 | +55% | 1 | 1 | 0% | 1,823 | 4,736 | +160% | 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 +23 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.