Skills — Coleção de Competências para Claude
andreiRS/skills
16 skills modulares para planejamento, spec, arquitetura, TDD e testes
Ver no GitHubConjunto de 16 skills reutilizáveis em TypeScript para estender capacidades do Claude e outras LLMs. Cobre todo ciclo de desenvolvimento: planejamento, especificação, arquitetura, implementação TDD, testes e documentação. Cada skill é uma instrução concisa e composável.
Instalação
npx convez-ai andreiRS/skills1 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
Casos de uso
- Conduzir entrevistas estruturadas e validar planos com perguntas críticas
- Gerar especificações enxutas e arquivos de arquitetura (ADRs)
- Implementar funcionalidades com TDD e orquestrar múltiplos agentes
- Testar aplicações em produção e executar testes exploratórios adversariais
- Quebrar mudanças em commits atômicos e gerar documentação HTML
Veredito

Excelente coleção de prompts de alta qualidade e bem testados para engenheiros que querem automação de processos de desenvolvimento completos. Abordagem opinionada mas flexível; valor real em orquestração TDD e análise arquitetônica.
README
A tradução para o português deste README ainda está na fila.
Skills
A personal collection of reusable skills for Claude Code and other LLM tools. The source for the banner above lives in assets/skills-map.html.
Skills
| Skill | Description |
|---|---|
interview | High-level interview to align on the shape of a plan (~3-6 questions) |
poke-holes | Focused review that hunts for weak spots and hidden assumptions (~6-12 questions) |
grill-me | Exhaustive walk of every branch of the decision tree, hands off to to-spec |
to-spec | Capture the why and what into a lean six-section spec under docs/specs/ (capture-first) |
domain-docs | Maintain the GLOSSARY.md glossary and docs/adr/ decision records |
improve-architecture | Find deepening opportunities (shallow→deep modules) and present them as a visual HTML report |
to-html | Render what's been discussed as a single self-contained HTML page — explainer, matrix, dashboard, cheat-sheet |
to-issues | Break a plan or spec into independently-grabbable issues as thin vertical slices |
implement-tdd | Build an existing plan test-first, one behavior at a time, red-green-refactor |
orchestrate-tdd | Drive a list of issues to completion by spawning one TDD agent per slice, sizing the model, verifying and reviewing each result |
prove-it | Drive the running app like a human to confirm the changed flows work end to end (pass/fail verdict) |
break-it | Adversarial exploratory testing — drive the running app to find the bugs a human QA would catch |
handoff | Summarise the current conversation into a handoff doc for the next session |
to-atomic-commits | Split uncommitted changes into logical atomic commits, respecting detected commit conventions |
garmin | Download Garmin Connect activity splits and summarize workouts |
brand-guidelines | Apply Andrei's personal brand colors, typography, and visual style to any rendered artifact |
Size at a glance
Each skill's SKILL.md — the instruction prose only — measured to keep us honest about conciseness. Tokens are estimated as bytes ÷ 4. Regenerate after any change with bun scripts/skill-sizes.ts --write.
| Skill | Lines | Words | ~Tokens |
|---|---|---|---|
orchestrate-tdd | 146 | 1,782 | 2,729 |
implement-tdd | 150 | 1,344 | 1,965 |
show-it | 60 | 1,032 | 1,561 |
to-html | 62 | 961 | 1,483 |
improve-architecture | 69 | 889 | 1,428 |
domain-docs | 105 | 872 | 1,409 |
to-spec | 105 | 878 | 1,401 |
garmin | 76 | 761 | 1,268 |
to-issues | 102 | 728 | 1,132 |
to-atomic-commits | 81 | 591 | 937 |
brand-guidelines | 67 | 396 | 679 |
break-it | 37 | 386 | 619 |
grill-me | 31 | 359 | 567 |
poke-holes | 37 | 321 | 503 |
prove-it | 29 | 328 | 493 |
interview | 27 | 228 | 367 |
handoff | 24 | 173 | 257 |
| Total | 1,208 | 12,029 | 18,798 |
Requirements
Most skills are prose-only and need nothing beyond Claude. These few shell out to external tools — install them before use:
| Tool | Used by | Install |
|---|---|---|
bun | garmin, show-it, installing agent-browser | bun.sh |
agent-browser | garmin, prove-it, break-it | bun add -g agent-browser && agent-browser install |
| Google Chrome (real app) | garmin | google.com/chrome |
garmin drives real Chrome with a persistent authenticated profile (to pass Garmin's Cloudflare bot check) and always runs headed — see its SKILL.md for the one-time sign-in.
From idea to shipped code
The plan-and-build skills chain together. Pick the review depth that fits, turn the result into work, then implement it:
- Shape the plan — three depth tiers for stress-testing a plan, pick based on how deep you want to go:
- interview → quick alignment, big-picture only
- poke-holes → probe the 2-3 riskiest parts
- grill-me → exhaustive, then hands off to
to-spec
- Capture it —
to-specstructures the resolved plan into a lean six-section spec underdocs/specs/. Capture-first: it records what's been decided, asking only for missing required pieces. - Break it down —
to-issuesslices the spec into independently-grabbable issues. - Build it —
implement-tddconsumes the spec or issues and builds them test-first. - Check it in the real app — once it's built and the suite is green, drive the running app like a human: prove-it confirms the changed flows work end to end; break-it throws nasty inputs and edge cases at them to find what a human QA would catch. Distinct from the test suite (
implement-tdd/orchestrate-tdd) — this is the manual, browser-driven pass.
For a batch of issues you'd rather not build by hand, orchestrate-tdd sits one level above implement-tdd: instead of coding, it spawns one agent per slice (running implement-tdd on that slice), sizes each to the cheapest model that fits, then independently re-runs the tests and reviews the diff before advancing. Reviews are tier-gated (skip trivial, simplify for simple, code-review for complex, plus a final cross-slice pass), failures retry then escalate to the most-capable model, and it writes status back to the tracker as each slice lands. Use it to drive a whole to-issues backlog to a reviewed branch autonomously.
domain-docs runs alongside this flow as the keeper of durable domain knowledge: the GLOSSARY.md glossary and docs/adr/ decision records. grill-me and to-spec flag fuzzy terms and hard-to-reverse decisions and point here to record them; it also works standalone.
improve-architecture feeds the front of this flow for existing codebases: it surveys the code, surfaces the strongest deepening opportunities as a visual HTML report, and stops. Take a chosen candidate into grill-me to design it, or to-issues to slice it.
handoff supports this flow across sessions: when work spans multiple sessions or collaborators, write a handoff doc so a fresh session can pick up where the last one left off.
Other
- garmin → pull recent Garmin Connect activity data and summarize it.
Inspiration
This collection draws on the ideas, structure, and conventions of:
- Anthropic — the official skills repo: anthropics/skills
- Matt Pocock — mattpocock/skills