From capability-fabric
Routes a task to the smallest set of skills that covers it and sequences them without context bloat, instead of one monolithic instruction. Use when a task spans multiple capabilities, when many skills are installed, or when deciding which skills to apply. Reads the skill registry to discover specialists and respects the tier/dependency ordering.
How this skill is triggered — by the user, by Claude, or both
Slash command
/capability-fabric:composing-skillsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
A library of small, sharp skills beats one giant skill, but only if the right ones are
A library of small, sharp skills beats one giant skill, but only if the right ones are selected and sequenced per task. This skill is the router and conductor: it turns "I have 100 skills" into "I'm using these 3, in this order, right now."
- [ ] 1 CLASSIFY the task: what capabilities does it actually need?
- [ ] 2 MATCH candidate skills by description (+ registry for non-resident specialists)
- [ ] 3 MINIMIZE pick the SMALLEST covering set; prefer the most specific skill per need
- [ ] 4 ORDER sequence by the dependency graph (tier N may use tier <=N)
- [ ] 5 LOAD activate one body at a time, as each phase begins (not all upfront)
Follow the tier/dependency rules (see docs/06):
decomposing-tasks → (sustained-execution if long) → domain work → verifying-reasoning.progressive-research before/within planning when facts are
missing.context-budgeting is ambient, not a phase.operating-autonomously is the outer loop for
unattended runs.sustained-execution ledger, not by stuffing everything in the window.Keep skills/REGISTRY.md as a one-line-per-skill index (name, when-to-use). Frequently
used skills sit in the resting menu; rare specialists live registry-only and are discovered
by reading the registry on demand. This caps resting token cost at O(hot set), not
O(all skills). See docs/design/07-activation-strategy.md.
| Task | Covering set (ordered) |
|---|---|
| "Research X and write a report" | progressive-research → verifying-reasoning (+budget) |
| "Build feature over a long session" | decomposing-tasks → sustained-execution → guarding-code-quality → verifying-reasoning |
| "Run this goal unattended" | operating-autonomously ⟨ wraps the build chain ⟩ |
| "Make me a skill for Y" | authoring-skills → composing-skills(sizing) → auditing-skills |
$name): this skill names which skills to invoke
and in what order.references.md, routing heuristics, overlap resolution, anti-monolith.examples.md, routing traces for real tasks.templates/composition-plan.md, record the chosen set + order.checklists/composition.md, pre-execution routing gate.benchmarks/, routing precision/recall & token-vs-monolith method.Provides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
npx claudepluginhub adrianparedez/capability-fabric --plugin capability-fabric