Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when tailoring content formats, hooks, and publishing mechanics for each social platform.
.claude/skills/gtmagents-platform-frameworks/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 5% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -13% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -38% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | -5% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | -7% | 0% |
| Platform | Cadence | Format Tips | CTA Guidance | | --- | --- | --- | --- | | LinkedIn | 3-5x/week | Lead with insight, use whitespace, add doc carousel | Invite comments + resource download | | X (Twitter) | Daily | Hooks < 50 chars, thread sequencing, emoji sparingly | Ask for RT/quote or drive to thread | | TikTok/Reels | 3-4x/week | Hook in 2s, captions, native text, trending audio | Soft CTA spoken + on-screen | | Instagram | 4-5x/week | Carousel storytelling, Reels for reach, alt text | DM or link-in-bio prompts |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 11,466 | 19,614 | +71% | 1 | 1 | 0% | 1,700 | 1,623 | -5% | 0 | 0 | — |
case-02 | pass→pass | 13,775 | 10,737 | -22% | 1 | 1 | 0% | 2,100 | 1,954 | -7% | 0 | 0 | — |
case-03 | fail→pass | 7,961 | 5,668 | -29% | 1 | 1 | 0% | 1,295 | 1,365 | +5% | 0 | 0 | — |
case-04 | pass→pass | 13,960 | 10,747 | -23% | 1 | 1 | 0% | 2,125 | 1,861 | -12% | 0 | 0 | — |
case-05 | pass→pass | 10,524 | 4,830 | -54% | 1 | 1 | 0% | 1,599 | 1,109 | -31% | 0 | 0 | — |
case-06 | pass→pass | 11,397 | 8,302 | -27% | 1 | 1 | 0% | 1,684 | 1,609 | -4% | 0 | 0 | — |
case-07 | fail→pass | 13,879 | 9,729 | -30% | 1 | 1 | 0% | 2,129 | 1,847 | -13% | 0 | 0 | — |
case-08 | pass→pass | 7,100 | 5,217 | -27% | 1 | 1 | 0% | 1,093 | 1,141 | +4% | 0 | 0 | — |
case-09 | fail→pass | 11,009 | 4,551 | -59% | 1 | 1 | 0% | 1,698 | 1,060 | -38% | 0 | 0 | — |
case-10 | pass→pass | 12,109 | 6,983 | -42% | 1 | 1 | 0% | 1,898 | 1,403 | -26% | 0 | 0 | — |
case-11 | pass→pass | 14,478 | 6,034 | -58% | 1 | 1 | 0% | 2,162 | 1,219 | -44% | 0 | 0 | — |
case-12 | pass→pass | 14,635 | 13,185 | -10% | 1 | 1 | 0% | 2,219 | 2,230 | +0% | 0 | 0 | — |
case-13 | pass→pass | 16,388 | 4,339 | -74% | 1 | 1 | 0% | 2,355 | 1,020 | -57% | 0 | 0 | — |
case-14 | pass→pass | 13,896 | 8,486 | -39% | 1 | 1 | 0% | 1,882 | 1,582 | -16% | 0 | 0 | — |
case-15 | pass→pass | 17,862 | 10,184 | -43% | 1 | 1 | 0% | 2,354 | 1,756 | -25% | 0 | 0 | — |
case-16 | pass→pass | 16,025 | 2,216 | -86% | 1 | 1 | 0% | 2,171 | 705 | -68% | 0 | 0 | — |
case-17 | pass→pass | 12,044 | 5,958 | -51% | 1 | 1 | 0% | 1,686 | 1,195 | -29% | 0 | 0 | — |
case-18 | pass→pass | 16,227 | 11,793 | -27% | 1 | 1 | 0% | 2,203 | 1,911 | -13% | 0 | 0 | — |
case-19 | pass→pass | 12,754 | 9,990 | -22% | 1 | 1 | 0% | 1,884 | 1,825 | -3% | 0 | 0 | — |
case-20 | pass→pass | 14,771 | 14,103 | -5% | 1 | 1 | 0% | 2,231 | 2,591 | +16% | 0 | 0 | — |
case-21 | pass→pass | 18,717 | 14,403 | -23% | 1 | 1 | 0% | 2,763 | 2,560 | -7% | 0 | 0 | — |
case-22 | pass→pass | 20,603 | 15,253 | -26% | 1 | 1 | 0% | 3,313 | 2,959 | -11% | 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.