convez.ai
Skill
miovrag

claude-skills

miovrag/claude-skills

Ver no GitHub
InicialPluginScore · de 0 a 100

Claude Code skills and commands

Instalação

$npx convez-ai miovrag/claude-skills

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
1
Forks
0
Acompanhando
1
Saúde
28%
Último commit
há 2 meses
Versão
—
Issues abertas
0
Contribuidores
1

README

A tradução para o português deste README ainda está na fila.

Agent Skills

Single source of truth for all agent skills (Claude Code & friends). Lives at ~/.agents/skills/; ~/.claude/skills/ contains only symlinks into this directory. Invoke any skill as /<folder-name>.

Provenance

Own — CustomGPT.ai design work

SkillWhat it does
customgpt-designCustomGPT.ai design system, strict token mode (system root: ~/customgpt-design-system/)
pxoFull 13-step Product Experience OS workflow analysis
premortemStructured premortem risk analysis (8 steps)

Own — UX pattern library

Distilled from Vitaly Friedman's "Fixing Frustrating Design Patterns" workshop, Baymard, NN/g, and related sources.

SkillArea
filtering-uxFilter UIs — interaction, taxonomy, layout
sorting-uxSorting — filter-vs-sort, faceted sorting, labels
data-table-uxTables/lists/grids — layout, interaction, a11y
navigation-uxNav, breadcrumbs, menus, IA
web-forms-uxForms, validation, multi-step, a11y
dashboard-datavizDashboards, chart selection, KPI cards
search-and-content-uxSearch, autocomplete, scanability
voice-uxVoice interfaces, conversation design
ai-design-patternsHuman-AI interaction, trust, transparency
prompt-engineeringPrompt design and refinement for LLMs

The list-view trio (filtering-ux, sorting-ux, data-table-ux) is designed to load together whenever a list/table/grid is in scope.

Own — analysis frameworks

SkillWhat it does
bourdieu-distinctionBourdieu's Distinction framework → 8-section sociological analysis

Third-party (vendored)

SetSourceSkills
Layers of Product Designjamiemill/layers-skillslayers-intro, layers-orient, layers-domain, layers-observed-behaviour, layers-user-needs, layers-product-strategy, layers-conceptual-model, layers-interaction-flow, layers-surface
Accessibility (Mike Gifford / CivicActions)mgifford/accessibility-skills27 granular a11y skills: accessibility-general, color-contrast, keyboard, forms, tables, tooltips, navigation, aria-live-regions, image-alt-text, svg, charts-graphs, touch-pointer, light-dark-mode, plain-language, content-design, manual-testing, axe-rules, bug-reporting, ci-cd, audio-video, maps, mermaid, print, progressive-enhancement, user-personalization, anchor-links, opquast-digital-quality
Remotionremotion-dev via npx skillsremotion-best-practices
skills.shvercel-labs/skillsfind-skills, project-kickstart
Designer skillsOwl-Listener/designer-skills~93 skills across design-research, design-systems, interaction-design, design-ops, designer-toolkit (critiques, UX laws, tokens, research methods, specs)
PM skillsphuryn/pm-skills68 skills across product-discovery, product-strategy, execution, data-analytics, market-research, go-to-market, marketing-growth, ai-shipping, toolkit
Animatedelphi-ai/animate-skillanimate — React/Next.js animation patterns (Emil Kowalski's course)
Journey Mappingcuellarfr/design-skillsjourney-mapping — alignment diagrams (CJM, service blueprints, experience/empathy maps, JJG Visual Vocabulary) with references/templates/examples; single skill vendored from the set to avoid name collisions

Name collisions: interview-script and summarize-interview exist in both designer-skills and pm-skills — the pm-skills versions (installed later, Mom Test/JTBD-based) are the ones on disk.

Third-party sets are vendored copies — re-installable via npx skills add <source>, but tracked here so local modifications survive.

Which analysis skill when

  • /pxo — full product-experience flow analysis of an existing screen/flow
  • /premortem — pre-launch risk list ("it already failed — why?")
  • /layers-orient — diagnose which design layer (research → strategy → model → flow → surface) is the bottleneck; then drill in with the matching /layers-* skill (load /layers-intro first)
Ver o original no GitHub

Bom para

Relacionados