Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Claude Skills meta-skill: extract domain material (docs/APIs/code/specs) into a reusable Skill (SKILL.md + references/scripts/assets), and refactor existing Skills for clarity, activation reliability, and quality gates.
.claude/skills/foryourhealth111-pixel-claude-skills/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 43% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 268% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 83% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 88% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 95% | 0% |
Turn scattered domain material into a Skill that is reusable, maintainable, and reliably activatable:
SKILL.md as the entrypoint (triggers, constraints, patterns, examples)references/ for long-form evidence and navigationscripts/ and assets/ for scaffolding and templatesTrigger this meta-skill when you need to:
references/This meta-skill is NOT:
Your output MUST include:
skills/<skill-name>/)SKILL.md with decidable triggers, boundaries, and reproducible examplesreferences/ with a references/index.mdskill-name/
|-- SKILL.md # Required: entrypoint with YAML frontmatter
|-- references/ # Optional: long-form docs/evidence/index
| `-- index.md # Recommended: navigation index
|-- scripts/ # Optional: helpers/automation
`-- assets/ # Optional: templates/configs/static assetsThe truly minimal version is just SKILL.md (you can add references/ later).
yaml--- name: skill-name description: "What it does + when to use (activation triggers)." ---
Frontmatter rules:
name MUST match ^[a-z][a-z0-9-]*$ and SHOULD match the directory namedescription MUST be decidable (not "helps with X") and include concrete trigger keywordsSKILL.md Skeleton (Copy/Paste)markdown--- name: my-skill description: "[Domain] capability: includes [capability 1], [capability 2]. Use when [decidable triggers]." --- # my-skill Skill One sentence that states the boundary and the deliverable. ## When to Use This Skill Trigger when any of these applies: - [Trigger 1: concrete task/keyword] - [Trigger 2] - [Trigger 3] ## Not For / Boundaries - What this skill will not do (prevents misfires and over-promising) - Required inputs; ask 1-3 questions if missing ## Quick Reference ### Common Patterns **Pattern 1:** one-line explanation
command/snippet you can paste and run]
## Examples
### Example 1
- Input:
- Steps:
- Expected output / acceptance:
### Example 2
### Example 3
## References
- `references/index.md`: navigation
- `references/...`: long-form docs split by topic
## Maintenance
- Sources: docs/repos/specs (do not invent)
- Last updated: YYYY-MM-DD
- Known limits: what is explicitly out of scopereferences/.description should say "what + when" with concrete keywords.Do not skip steps:
references/ + write references/index.mdMinimum checks (see references/quality-checklist.md for the full version):
name matches ^[a-z][a-z0-9-]*$ and matches the directory namedescription states "what + when" with concrete trigger keywordsreferences/ and references/index.md is navigableValidate locally:
bash# From repo root (basic validation) ./skills/claude-skills/scripts/validate-skill.sh skills/<skill-name> # From repo root (strict validation) ./skills/claude-skills/scripts/validate-skill.sh skills/<skill-name> --strict # From skills/claude-skills/ (basic validation) ./scripts/validate-skill.sh ../<skill-name> # From skills/claude-skills/ (strict validation) ./scripts/validate-skill.sh ../<skill-name> --strict
Generate a new Skill skeleton:
bash# From repo root (generate into ./skills/) ./skills/claude-skills/scripts/create-skill.sh my-skill --full --output skills # From skills/claude-skills/ (generate into ../ i.e. ./skills/) ./scripts/create-skill.sh my-skill --full --output .. # Minimal skeleton ./skills/claude-skills/scripts/create-skill.sh my-skill --minimal --output skills
Templates:
assets/template-minimal.mdassets/template-complete.mdcreate-skill.sh to scaffold skills/<skill-name>/description as "what + when"references/ and wire references/index.mdvalidate-skill.sh --strict and iterateSKILL.md with long pasted documentationreferences/ (split by topic)skills/<skill-name>/validate-skill.sh (non-strict) to get warningsvalidate-skill.sh --strict to enforce the specreferences/quality-checklist.md before shippingLocal docs:
references/index.mdreferences/skill-spec.mdreferences/quality-checklist.mdreferences/anti-patterns.mdreferences/README.md (upstream official reference)External (official):
skills/claude-skills/references/ + upstream official docs in references/README.mdvalidate-skill.sh is heuristic; strict mode assumes the recommended section headings| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 21,979 | 22,449 | +2% | 1 | 1 | 0% | 4,466 | 6,382 | +43% | 0 | 0 | — |
case-02 | pass→pass | 30,064 | 65,151 | +117% | 1 | 1 | 0% | 5,407 | 6,899 | +28% | 0 | 0 | — |
case-03 | fail→fail | 4,664 | 4,289 | -8% | 1 | 1 | 0% | 285 | 2,423 | +750% | 0 | 0 | — |
case-04 | pass→pass | 10,405 | 7,678 | -26% | 1 | 1 | 0% | 2,271 | 3,940 | +73% | 0 | 0 | — |
case-05 | fail→pass | 4,724 | 5,510 | +17% | 1 | 1 | 0% | 872 | 3,211 | +268% | 0 | 0 | — |
case-06 | pass→pass | 11,996 | 11,695 | -3% | 1 | 1 | 0% | 2,487 | 4,512 | +81% | 0 | 0 | — |
case-07 | pass→pass | 4,839 | 5,870 | +21% | 1 | 1 | 0% | 882 | 2,871 | +226% | 0 | 0 | — |
case-08 | pass→pass | 9,131 | 5,772 | -37% | 1 | 1 | 0% | 1,430 | 3,191 | +123% | 0 | 0 | — |
case-09 | pass→pass | 14,283 | 11,274 | -21% | 1 | 1 | 0% | 2,553 | 4,258 | +67% | 0 | 0 | — |
case-10 | pass→pass | 10,488 | 8,316 | -21% | 1 | 1 | 0% | 1,759 | 3,612 | +105% | 0 | 0 | — |
case-11 | pass→pass | 9,608 | 6,154 | -36% | 1 | 1 | 0% | 1,502 | 3,140 | +109% | 0 | 0 | — |
case-12 | fail→pass | 6,678 | 2,922 | -56% | 1 | 1 | 0% | 1,354 | 2,475 | +83% | 0 | 0 | — |
case-13 | fail→pass | 8,638 | 1,708 | -80% | 1 | 1 | 0% | 1,305 | 2,451 | +88% | 0 | 0 | — |
case-14 | fail→pass | 8,385 | 3,146 | -62% | 1 | 1 | 0% | 1,364 | 2,665 | +95% | 0 | 0 | — |
case-15 | fail→pass | 12,415 | 9,694 | -22% | 1 | 1 | 0% | 2,097 | 3,998 | +91% | 0 | 0 | — |
case-16 | fail→pass | 11,007 | 4,650 | -58% | 1 | 1 | 0% | 1,868 | 2,935 | +57% | 0 | 0 | — |
case-17 | pass→pass | 5,507 | 2,881 | -48% | 1 | 1 | 0% | 967 | 2,663 | +175% | 0 | 0 | — |
case-18 | pass→pass | 6,175 | 3,388 | -45% | 1 | 1 | 0% | 1,192 | 2,750 | +131% | 0 | 0 | — |
case-19 | pass→pass | 10,667 | 5,976 | -44% | 1 | 1 | 0% | 1,750 | 3,126 | +79% | 0 | 0 | — |
case-20 | pass→pass | 8,337 | 6,557 | -21% | 1 | 1 | 0% | 1,428 | 3,163 | +121% | 0 | 0 | — |
case-21 | fail→pass | 23,743 | 2,093 | -91% | 1 | 1 | 0% | 4,287 | 2,496 | -42% | 0 | 0 | — |
case-22 | pass→pass | 12,825 | 9,447 | -26% | 1 | 1 | 0% | 2,192 | 3,849 | +76% | 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, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +36 percentage points is the difference between those two pass rates over the 21 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.