Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Operational template for building social content calendars with approvals and publishing workflows.
.claude/skills/gtmagents-social-calendar-system/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-06 | ✗→✓ | ▲ Improved | -5% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -35% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 37% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -8% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 3% | 0% |
build-social-calendar for auto-populated grids.Week Ending: <Date>
Channel | Goal | Target KPI | Actual | Guardrail | Notes
LinkedIn | Demand gen POV | Engagement ≥2.5% | 3.1% | 1.8% | Exceeded; repurpose for email snippet
X | Thought leadership | CTR ≥1.5% | 1.1% | 1.0% | Slight dip; test hook variant
...
Top Performing Assets:
- <Title> – reason, reuse plan
Risks/Escalations:
- <Issue> – owner, action| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | pass→pass | 23,812 | 34,650 | +46% | 1 | 1 | 0% | 4,073 | 4,583 | +13% | 0 | 0 | — |
case-02 | fail→fail | 19,335 | 13,843 | -28% | 1 | 1 | 0% | 2,867 | 2,879 | +0% | 0 | 0 | — |
case-03 | pass→pass | 18,359 | 19,891 | +8% | 1 | 1 | 0% | 2,707 | 3,736 | +38% | 0 | 0 | — |
case-04 | pass→pass | 16,786 | 16,483 | -2% | 1 | 1 | 0% | 2,639 | 3,272 | +24% | 0 | 0 | — |
case-05 | pass→pass | 22,221 | 23,145 | +4% | 1 | 1 | 0% | 3,037 | 3,882 | +28% | 0 | 0 | — |
case-06 | fail→pass | 13,274 | 7,229 | -46% | 1 | 1 | 0% | 1,966 | 1,862 | -5% | 0 | 0 | — |
case-07 | fail→pass | 14,300 | 4,772 | -67% | 1 | 1 | 0% | 2,238 | 1,454 | -35% | 0 | 0 | — |
case-08 | fail→pass | 17,606 | 17,833 | +1% | 1 | 1 | 0% | 2,372 | 3,242 | +37% | 0 | 0 | — |
case-09 | pass→pass | 11,987 | 12,316 | +3% | 1 | 1 | 0% | 1,796 | 2,364 | +32% | 0 | 0 | — |
case-10 | pass→pass | 11,044 | 5,831 | -47% | 1 | 1 | 0% | 1,621 | 1,476 | -9% | 0 | 0 | — |
case-11 | fail→pass | 12,722 | 6,448 | -49% | 1 | 1 | 0% | 1,840 | 1,697 | -8% | 0 | 0 | — |
case-12 | pass→pass | 12,842 | 2,853 | -78% | 1 | 1 | 0% | 1,877 | 1,104 | -41% | 0 | 0 | — |
case-13 | fail→pass | 15,466 | 10,033 | -35% | 1 | 1 | 0% | 2,082 | 2,153 | +3% | 0 | 0 | — |
case-14 | fail→fail | 14,237 | 11,631 | -18% | 1 | 1 | 0% | 2,113 | 2,254 | +7% | 0 | 0 | — |
case-15 | fail→pass | 11,534 | 10,928 | -5% | 1 | 1 | 0% | 1,610 | 2,170 | +35% | 0 | 0 | — |
case-24 | pass→pass | 19,987 | 17,484 | -13% | 1 | 1 | 0% | 2,854 | 3,233 | +13% | 0 | 0 | — |
case-16 | pass→pass | 12,543 | 13,069 | +4% | 1 | 1 | 0% | 1,631 | 2,399 | +47% | 0 | 0 | — |
case-17 | pass→pass | 12,545 | 12,276 | -2% | 1 | 1 | 0% | 1,788 | 2,340 | +31% | 0 | 0 | — |
case-18 | pass→pass | 5,049 | 4,429 | -12% | 1 | 1 | 0% | 720 | 1,266 | +76% | 0 | 0 | — |
case-19 | pass→pass | 10,077 | 5,261 | -48% | 1 | 1 | 0% | 1,414 | 1,481 | +5% | 0 | 0 | — |
case-20 | pass→pass | 9,234 | 4,442 | -52% | 1 | 1 | 0% | 1,525 | 1,337 | -12% | 0 | 0 | — |
case-21 | pass→pass | 12,446 | 6,058 | -51% | 1 | 1 | 0% | 1,771 | 1,565 | -12% | 0 | 0 | — |
case-22 | fail→pass | 11,230 | 2,439 | -78% | 1 | 1 | 0% | 1,594 | 1,068 | -33% | 0 | 0 | — |
case-23 | pass→pass | 17,092 | 14,805 | -13% | 1 | 1 | 0% | 2,455 | 2,860 | +16% | 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 +29 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.