Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →An example user-invoked skill that demonstrates frontmatter options and the skills/<name>/SKILL.md layout
.claude/skills/anthropics-example-command/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | -70% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -37% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -83% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -66% | 0% |
| case-03 | ✓→✓ | ＝ Same ✓ | -43% | 0% |
This demonstrates the skills/<name>/SKILL.md layout for user-invoked slash commands. It is functionally identical to the legacy commands/example-command.md format — both are loaded the same way; only the file layout differs.
The user invoked this with: $ARGUMENTS
When this skill is invoked:
Skills in this layout support these frontmatter fields:
/example-command my-argument
/example-command arg1 arg2| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 7,541 | 1,853 | -75% | 1 | 1 | 0% | 1,304 | 495 | -62% | 0 | 0 | — |
case-02 | fail→fail | 5,826 | 3,841 | -34% | 1 | 1 | 0% | 1,053 | 883 | -16% | 0 | 0 | — |
case-03 | pass→pass | 8,066 | 3,335 | -59% | 1 | 1 | 0% | 1,443 | 825 | -43% | 0 | 0 | — |
case-04 | fail→pass | 9,072 | 1,929 | -79% | 1 | 1 | 0% | 1,846 | 559 | -70% | 0 | 0 | — |
case-05 | pass→pass | 9,058 | 3,438 | -62% | 1 | 1 | 0% | 1,531 | 874 | -43% | 0 | 0 | — |
case-06 | fail→pass | 6,601 | 2,571 | -61% | 1 | 1 | 0% | 1,026 | 645 | -37% | 0 | 0 | — |
case-07 | pass→pass | 5,444 | 1,551 | -72% | 1 | 1 | 0% | 908 | 517 | -43% | 0 | 0 | — |
case-08 | fail→pass | 16,183 | 1,872 | -88% | 1 | 1 | 0% | 2,674 | 447 | -83% | 0 | 0 | — |
case-09 | pass→pass | 5,574 | 5,018 | -10% | 1 | 1 | 0% | 951 | 495 | -48% | 0 | 0 | — |
case-14 | pass→pass | 5,197 | 1,629 | -69% | 1 | 1 | 0% | 843 | 462 | -45% | 0 | 0 | — |
case-10 | pass→pass | 6,174 | 2,134 | -65% | 1 | 1 | 0% | 1,005 | 507 | -50% | 0 | 0 | — |
case-11 | pass→pass | 7,440 | 1,608 | -78% | 1 | 1 | 0% | 1,280 | 465 | -64% | 0 | 0 | — |
case-12 | pass→pass | 5,359 | 1,981 | -63% | 1 | 1 | 0% | 850 | 509 | -40% | 0 | 0 | — |
case-13 | fail→pass | 9,911 | 2,073 | -79% | 1 | 1 | 0% | 1,729 | 581 | -66% | 0 | 0 | — |
case-15 | pass→pass | 14,722 | 6,730 | -54% | 1 | 1 | 0% | 2,710 | 1,440 | -47% | 0 | 0 | — |
case-16 | pass→pass | 4,452 | 1,770 | -60% | 1 | 1 | 0% | 822 | 483 | -41% | 0 | 0 | — |
case-17 | pass→pass | 4,642 | 1,697 | -63% | 1 | 1 | 0% | 876 | 443 | -49% | 0 | 0 | — |
case-18 | pass→pass | 5,732 | 1,677 | -71% | 1 | 1 | 0% | 1,103 | 533 | -52% | 0 | 0 | — |
case-19 | pass→pass | 7,677 | 2,324 | -70% | 1 | 1 | 0% | 1,446 | 646 | -55% | 0 | 0 | — |
case-20 | pass→pass | 10,671 | 8,670 | -19% | 1 | 1 | 0% | 2,052 | 2,068 | +1% | 0 | 0 | — |
case-21 | pass→pass | 9,132 | 9,218 | +1% | 1 | 1 | 0% | 1,681 | 2,000 | +19% | 0 | 0 | — |
case-22 | pass→pass | 10,343 | 8,619 | -17% | 1 | 1 | 0% | 1,888 | 1,756 | -7% | 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 +18 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.