agentleFS
felvieirafelvieira/claude-skills-fv/skills/44-zoom-out/SKILL.md

zoom-out

Skill28 starsChanged 3 months agoApache-2.0View on GitHub

Mapeia modulos, chamadas e relacoes para entender o sistema. Trigger em: "zoom out", "mapa de modulos", "mapa dos modulos", "estou perdido", "perdido nesse codigo", "como isso encaixa", "como se conecta", "callers principais", "visao geral", "big picture", "neighborhood", "broader context", "higher level perspective".

118 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use to-prd, to-issues and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • GoogleSkill

    to-prd

    Turn the current conversation into a PRD and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

    Also in 62 projects that use this

  • GoogleSkill

    to-issues

    Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices.

    Also in 50 projects that use this

  • simo97Skill

    request-refactor-plan

    Use this skill when user wants to create a refactor plan.

    Also in 17 projects that use this

  • simo97Skill

    design-an-interface

    Generate multiple radically different interface designs for a module using parallel sub-agents. Use when user wants to design an API, explore interface options, compare module shapes, or mentions "design it twice".

    Also in 17 projects that use this

  • kennethkeimSkill

    ubiquitous-language

    Extract a DDD-style ubiquitous language glossary from the current conversation, flagging ambiguities and proposing canonical terms. Saves to UBIQUITOUS_LANGUAGE.md. Use when user wants to define domain terms, build a glossary, harden terminology, create a ubiquitous language, or mentions "domain model" or "DDD".

    Also in 16 projects that use this

  • Code-ShockSkill

    git-guardrails

    Set up git guardrails — a Claude Code PreToolUse hook that intercepts and blocks dangerous git commands such as force-push and hard reset before they run. Use to install git safety hooks or when the user runs /git-guardrails.

    Also in 3 projects that use this

  • +3 more from simo97
  • +1 more from google

The file

Files
SKILL.md

Zoom Out — Mapa de Bairro Antes do Codigo

Inspiracao: mattpocock/skills/engineering/zoom-out (MIT). Adaptado: forca uso de graphify-out/graph.json antes de Grep/Read brutos (CLAUDE.md global).

Quando Usar

  • ao receber task em modulo que o agente nao conhece bem
  • quando o usuario disse "estou perdido nessa parte"
  • antes de propor refactor ou architecture change (input pra skill 38)
  • antes de explorar com Grep ou Read direto (mais economico)
  • como preludio de detective-spec em codigos legados

Quando NAO Usar

  • voce ja conhece o modulo (zoom out vira ruido)
  • task pontual em arquivo unico ja identificado
  • bug fix com stack trace claro

Governanca Global

Esta skill segue GLOBAL.md, policies/code-exploration.md (graph > grep > read), policies/token-efficiency.md (mapa enxuto, nao dump), policies/handoffs.md (o mapa pode ser consumido por outras skills).

Protocolo

1. Tentar graph primeiro

Antes de qualquer Read/Grep, tentar:

test -f graphify-out/graph.json && echo "graph disponivel" || echo "sem graph"

Se graph existir, ler graphify-out/graph.json + graphify-out/GRAPH_REPORT.md (god nodes, communities). Isso responde 80% das perguntas de mapa.

2. Fallback: descoberta estrutural

Sem graph, descobrir estrutura via Glob:

# Topologia
find <alvo> -type f \( -name "*.ts" -o -name "*.tsx" -o -name "*.js" -o -name "*.py" -o -name "*.go" -o -name "*.rs" \) | head -50

Identificar:

  • entry points (index., main., app., cli.)
  • agrupamentos por pasta (modulos logicos)
  • arquivos grandes (provaveis hubs)

3. Mapear callers/callees principais

Para cada hub identificado, Grep por importacoes:

# Quem usa o modulo X?
rg "from ['\"].*<modulo-X>" --type ts --type tsx -l

4. Glossario do projeto

Antes de produzir output, capturar vocabulario do dominio lendo (em ordem de preferencia):

  1. memory/constitution.md (se existir — termos canonicos)
  2. docs/repo-audit/current.md (audit)
  3. README.md (sintese)
  4. Nomes de pastas/arquivos (fallback)

O mapa fala a lingua do projeto, nao termos genericos.

Checkpoint antes de entregar o mapa: para cada "hub" ou "god node" listado, confirmar com um segundo grep (contagem real de importações, não só a posição no graph.json) — graph desatualizado pode listar um arquivo já removido/refatorado como hub. Item que não confirma no código atual sai do mapa ou vira nota "desatualizado desde X".

Output

Tabela markdown ou hierarquia textual:

# Mapa de <area>

## Vocabulario do dominio
- <termo>: <definicao curta>

## Arquitetura geral
<diagrama ASCII opcional — so se reduz complexidade>

## Modulos principais

| Modulo | Proposito | Callers principais | Callees principais |
|--------|-----------|-------------------|-------------------|
| ...    | ...       | ...               | ...               |

## God nodes (>X dependentes)
- <arquivo>: <numero> dependentes — provavel ponto de friccao

## Pendencias visiveis
- <TODO/FIXME publicos>
- <gaps documentacao>

Handoffs

  • input pra skill 38 (architecture-deepener): mapa identifica god nodes; skill 38 propoe deepening
  • input pra skill 33 (detective-spec): mapa orienta quais modulos detective deve cavar primeiro
  • input pra skill 09 (orchestrator): mapa informa onde a feature toca, qual pipeline montar
  • input pra skill 32 (smart-suggestions): dado o mapa, sugerir proxima acao concreta

Anti-padroes

  • ❌ Dump completo de find . -name "*.ts" — isso e fuga, nao mapa
  • ❌ Ler 50 arquivos pra entender 1 modulo — usa graph primeiro
  • ❌ Mapa generico sem vocabulario do projeto
  • ❌ "vou ler tudo e depois mapeio" — produz o mapa enquanto explora, nao depois

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.