Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Workflows

ASecurity

Use when orchestrating large parallel work or deciding between chat/skill/subagent/agent-team/workflow. Triggers on "workflow", "dynamic workflows", "agentes em paralelo", "força-tarefa", "dezenas de agentes", "escada de recursos", "/workflows", "/goal", "/loop". Keeps token costs predictable: scope, worker model, monitoring.

2 stars
0 votes
0 copies
0 views
Added 9/19/2026
ai-agentsgo

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

$npx -y skills add majinmagros/magros.ai-skills --skill workflows --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Workflows?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Workflows
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/majinmagros-workflows/badge)](https://www.skillsdirectory.com/skills/majinmagros-workflows)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: workflows
description: Use when orchestrating large parallel work or deciding between chat/skill/subagent/agent-team/workflow. Triggers on "workflow", "dynamic workflows", "agentes em paralelo", "força-tarefa", "dezenas de agentes", "escada de recursos", "/workflows", "/goal", "/loop". Keeps token costs predictable: scope, worker model, monitoring.
---

# Skill: Workflows — trabalho gigante em paralelo (dynamic workflows)

Decide QUANDO usar cada degrau da "escada de recursos" e roda workflows sem
surpresa na conta de tokens.

## Quando usar (gatilhos literais)

- "Roda isso em paralelo com vários agentes"
- "Chat, skill, subagente ou workflow?"
- "Força-tarefa em dezenas de arquivos"
- "/workflows: quanto tá gastando?"
- "Workflow ou agent team?"

## Exemplo

```text
Tarefa: "otimizar imagens das 12 páginas" → 12 workers Haiku + síntese Opus
Escopo fechado antes de disparar; monitora em /workflows
```

## 1. A escada (suba só até onde o problema existe)

| Degrau | O que é | Quando usar |
|---|---|---|
| **Chat direto** | 1 tarefa, 1 resposta | A maioria dos casos |
| **Skill** | Receita reaproveitável ("o como") | Tarefa que se repete igual |
| **Subagente** | Paralelo sem poluir a conversa; responde só ao central | Pesquisas paralelas independentes |
| **Agent team** | Equipe que DEBATE entre si (caro) | Resultado precisa de discussão |
| **Workflow** | Dezenas/centenas de agentes, script salvo e reexecutável | Trabalho gigante em paralelo ("o quantos") |

Pergunta-chave: os agentes precisam conversar entre si? Subagentes/workflows
NÃO conversam (cada um em janela própria). Agent team sim.

## 2. Profundidade vs largura

- `/go`: **profundidade** — loop até ATINGIR CRITÉRIO (um modelo pequeno
  verifica a conclusão). "Continue refinando até passar de 9/10 no checklist."
- `/loop`: repete por **tempo** (ex.: checar deploy a cada 5 min).
- **Workflow**: **largura** — muitos agentes lado a lado, costurados no fim
  ("cavar poço" vs "arar o campo").
- `/deep research` é um workflow (agentes cruzam fontes, votam, só sobra fato checado).

## 3. Custo (onde 99% se ferra)

- Cada agente do workflow = uma chamada completa (janela própria, contexto do zero).
  Workflow de 50 agentes ≈ abrir 50 conversas de uma vez.
- **Escopo fechado**: "otimize só as imagens das 12 páginas de produto", não
  "melhore meu site". Pedido vago = explosão de tokens.
- **Modelo**: Haiku nos agentes trabalhadores; Opus só na síntese final.
- **Monitore**: `/workflows` mostra fases, agentes e tokens; interrompa se estourar.
- Erros comuns: usar workflow pra tudo, escopo aberto, não acompanhar.

## 4. Processo

1. Pergunte se a tarefa se quebra em pedaços independentes. Se não → não é workflow.
2. Feche o escopo e o entregável antes de disparar.
3. Rode: Haiku nos workers + Opus na síntese; salve o script no `.claude/` do
   projeto se a equipe for reutilizar.
4. Acompanhe com `/workflows`; entregue o resultado costurado com as fontes/agentes usados.

## 5. Regra

- Recurso novo ≠ obrigatório. Renomear 1 variável = chat; renomear em 300 arquivos = workflow.

Attribution

majinmagrosmajinmagros
View sourceMore from majinmagros →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode. Cuts token usage ~75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.

1023331 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

686011 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3331 votes

catchup

Recovers prior coding-agent session context by running `catchup <agent> --since-compact`, which extracts a clean summary of a previous Codex, Claude Code, Antigravity, OpenCode, or Pi Agent session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", or asks to recover/summarize a previous session before continuing. Do NOT use for the current conversation, git history, or any non-agent log.

611 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →