Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Provide reusable patterns for validation, error handling, scaffolding. Use for skill consistency.
.claude/skills/athola-shared-patterns/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 27% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -53% | 0% |
Reusable patterns and templates for skill and hook development.
abstract:skill-authoring)This skill provides shared patterns that are referenced by other skills in the abstract plugin. It follows DRY principles by centralizing common patterns.
See modules/validation-patterns.md for:
See modules/error-handling.md for:
See modules/testing-templates.md for:
See modules/workflow-patterns.md for:
Reference these patterns from other skills:
markdownFor validation patterns, see the `shared-patterns` skill's [validation-patterns](../shared-patterns/modules/validation-patterns.md) module.
Verification: Run the command with --help flag to verify availability.
validation-patterns.md, error-handling.md,testing-templates.md, workflow-patterns.md) exists under plugins/abstract/skills/shared-patterns/modules/.
../shared-patterns/modules/ and resolves to an existing file on disk.
skills/shared/modules/directory pattern; such references are surfaced as broken links.
duplicated from the modules it links.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 8,597 | 3,002 | -65% | 1 | 1 | 0% | 1,313 | 965 | -27% | 0 | 0 | — |
case-02 | fail→pass | 13,171 | 12,338 | -6% | 1 | 1 | 0% | 2,185 | 2,773 | +27% | 0 | 0 | — |
case-03 | fail→pass | 6,360 | 6,167 | -3% | 1 | 1 | 0% | 992 | 1,545 | +56% | 0 | 0 | — |
case-04 | fail→pass | 17,993 | 4,360 | -76% | 1 | 1 | 0% | 3,066 | 1,183 | -61% | 0 | 0 | — |
case-05 | fail→pass | 12,896 | 2,970 | -77% | 1 | 1 | 0% | 1,991 | 928 | -53% | 0 | 0 | — |
case-06 | fail→pass | 10,443 | 5,840 | -44% | 1 | 1 | 0% | 1,522 | 1,511 | -1% | 0 | 0 | — |
case-07 | pass→pass | 8,264 | 1,453 | -82% | 1 | 1 | 0% | 1,183 | 735 | -38% | 0 | 0 | — |
case-08 | pass→pass | 13,407 | 4,807 | -64% | 1 | 1 | 0% | 1,935 | 1,287 | -33% | 0 | 0 | — |
case-09 | fail→pass | 12,540 | 2,883 | -77% | 1 | 1 | 0% | 2,070 | 894 | -57% | 0 | 0 | — |
case-10 | fail→pass | 12,566 | 1,543 | -88% | 1 | 1 | 0% | 1,919 | 719 | -63% | 0 | 0 | — |
case-11 | fail→pass | 8,697 | 1,821 | -79% | 1 | 1 | 0% | 1,289 | 781 | -39% | 0 | 0 | — |
case-12 | fail→pass | 11,709 | 2,742 | -77% | 1 | 1 | 0% | 1,843 | 877 | -52% | 0 | 0 | — |
case-13 | fail→pass | 16,387 | 3,446 | -79% | 1 | 1 | 0% | 2,775 | 1,068 | -62% | 0 | 0 | — |
case-14 | fail→pass | 8,484 | 2,775 | -67% | 1 | 1 | 0% | 1,303 | 981 | -25% | 0 | 0 | — |
case-15 | fail→pass | 11,736 | 3,356 | -71% | 1 | 1 | 0% | 1,936 | 1,000 | -48% | 0 | 0 | — |
case-16 | pass→pass | 10,196 | 2,131 | -79% | 1 | 1 | 0% | 1,613 | 742 | -54% | 0 | 0 | — |
case-17 | fail→pass | 9,312 | 3,811 | -59% | 1 | 1 | 0% | 1,362 | 1,089 | -20% | 0 | 0 | — |
case-18 | fail→pass | 3,277 | 1,823 | -44% | 1 | 1 | 0% | 465 | 752 | +62% | 0 | 0 | — |
case-19 | pass→pass | 6,775 | 2,042 | -70% | 1 | 1 | 0% | 1,022 | 752 | -26% | 0 | 0 | — |
case-20 | fail→pass | 11,209 | 1,716 | -85% | 1 | 1 | 0% | 1,907 | 729 | -62% | 0 | 0 | — |
case-21 | fail→pass | 16,451 | 2,289 | -86% | 1 | 1 | 0% | 2,498 | 851 | -66% | 0 | 0 | — |
case-22 | fail→pass | 6,767 | 1,767 | -74% | 1 | 1 | 0% | 965 | 698 | -28% | 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 +82 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.