Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants help creating, scheduling, or optimizing social media content for LinkedIn, Twitter/X, Instagram, TikTok, Facebook, or other platforms. Also use when the user mentions 'LinkedIn post,' 'Twitter thread,' 'social media,' 'content calendar,' 'social scheduling,' 'engagement,' or 'viral content.' This skill covers content creation, repurposing, and platform-specific strategies.
.claude/skills/dokhacgiakhoa-social-content/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-07 | ✓→✓ | ＝ Same ✓ | 317% | 0% |
| case-08 | ✓→✓ | ＝ Same ✓ | 73% | 0% |
| case-09 | ✓→✓ | ＝ Same ✓ | 23% | 0% |
| case-10 | ✓→✓ | ＝ Same ✓ | 83% | 0% |
You are an expert social media strategist with direct access to a scheduling platform that publishes to all major social networks. Your goal is to help create engaging content that builds audience, drives engagement, and supports business goals.
Gather this context (ask if not provided):
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 23,712 | 23,478 | -1% | 1 | 1 | 0% | 3,815 | 4,120 | +8% | 0 | 0 | — |
case-02 | fail→pass | 26,201 | 13,797 | -47% | 1 | 1 | 0% | 3,445 | 2,432 | -29% | 0 | 0 | — |
case-03 | fail→fail | 23,223 | 27,350 | +18% | 1 | 1 | 0% | 3,218 | 4,786 | +49% | 0 | 0 | — |
case-04 | fail→fail | 16,115 | 14,766 | -8% | 1 | 1 | 0% | 1,868 | 2,640 | +41% | 0 | 0 | — |
case-05 | fail→fail | 18,134 | 23,944 | +32% | 1 | 1 | 0% | 2,100 | 4,263 | +103% | 0 | 0 | — |
case-06 | fail→fail | 24,385 | 20,612 | -15% | 1 | 1 | 0% | 3,127 | 3,630 | +16% | 0 | 0 | — |
case-07 | pass→pass | 3,015 | 10,418 | +246% | 1 | 1 | 0% | 472 | 1,966 | +317% | 0 | 0 | — |
case-08 | pass→pass | 14,820 | 21,656 | +46% | 1 | 1 | 0% | 2,102 | 3,628 | +73% | 0 | 0 | — |
case-09 | pass→pass | 16,136 | 8,025 | -50% | 1 | 1 | 0% | 1,940 | 2,388 | +23% | 0 | 0 | — |
case-10 | pass→pass | 12,743 | 13,951 | +9% | 1 | 1 | 0% | 1,293 | 2,370 | +83% | 0 | 0 | — |
case-11 | pass→pass | 19,204 | 20,749 | +8% | 1 | 1 | 0% | 2,913 | 3,604 | +24% | 0 | 0 | — |
case-12 | pass→pass | 15,610 | 14,979 | -4% | 1 | 1 | 0% | 1,757 | 2,647 | +51% | 0 | 0 | — |
case-13 | pass→pass | 20,354 | 18,725 | -8% | 1 | 1 | 0% | 2,576 | 3,251 | +26% | 0 | 0 | — |
case-14 | fail→fail | 21,023 | 17,229 | -18% | 1 | 1 | 0% | 2,589 | 4,011 | +55% | 0 | 0 | — |
case-15 | pass→pass | 17,649 | 17,822 | +1% | 1 | 1 | 0% | 2,078 | 3,218 | +55% | 0 | 0 | — |
case-16 | pass→pass | 18,718 | 14,303 | -24% | 1 | 1 | 0% | 2,126 | 3,264 | +54% | 0 | 0 | — |
case-17 | pass→pass | 9,620 | 14,031 | +46% | 1 | 1 | 0% | 1,609 | 2,580 | +60% | 0 | 0 | — |
case-18 | pass→pass | 18,200 | 10,468 | -42% | 1 | 1 | 0% | 2,362 | 2,782 | +18% | 0 | 0 | — |
case-19 | pass→pass | 23,349 | 18,818 | -19% | 1 | 1 | 0% | 2,999 | 3,157 | +5% | 0 | 0 | — |
case-20 | pass→pass | 19,851 | 18,394 | -7% | 1 | 1 | 0% | 2,421 | 3,207 | +32% | 0 | 0 | — |
case-21 | pass→pass | 15,314 | 18,326 | +20% | 1 | 1 | 0% | 2,877 | 3,387 | +18% | 0 | 0 | — |
case-22 | pass→pass | 31,277 | 22,088 | -29% | 1 | 1 | 0% | 5,045 | 4,186 | -17% | 0 | 0 | — |
case-23 | pass→pass | 17,443 | 12,914 | -26% | 1 | 1 | 0% | 2,557 | 2,585 | +1% | 0 | 0 | — |
case-24 | pass→pass | 18,422 | 15,950 | -13% | 1 | 1 | 0% | 2,284 | 3,402 | +49% | 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. 24 cases were attempted. The headline lift of +4 percentage points is the difference between those two pass rates over the 24 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.