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

Unlazy

ASecurity

Enforces completion discipline with Depth Tree + runnable gates ledger. Use when agent returns half-done, 80% report, or UI incomplete — triggers on "unlazy", "lazy ai", "preguicosa", "depth tree", "gates", "do not stop until done".

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

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

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

Installs into .claude/skills of the current project.

Are you the author of Unlazy?

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

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

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

Download Zip
Files
SKILL.md
---
name: unlazy
description: Enforces completion discipline with Depth Tree + runnable gates ledger. Use when agent returns half-done, 80% report, or UI incomplete — triggers on "unlazy", "lazy ai", "preguicosa", "depth tree", "gates", "do not stop until done".
---

# Unlazy — Anti-Preguiça com Gates Executáveis

> Baseado em `Leonxlnx/unlazy` (MIT) apresentado em `Maestros da IA — 8fAuXr8fQBo` (2026-08-31). Fonte oficial: https://github.com/Leonxlnx/unlazy

Toda tarefa substancial vira um **ledger de gates** antes de começar. Done é o ledger, não a sensação.

## Quando usar

- IA entregou tecnicamente mas incompleto (80%, sem interação, placeholder)
- Relatório/tabela/app com funcionalidade faltando, precisa provar completo
- Build/audit/refactor com 3+ layers ou pipelines paralelos
- Explícito: `unlazy`, `tree N`, `gates`, `do not stop until done`

## Quando NÃO usar

- Edição trivial, resposta factual curta → use direto
- Precisa de gate visual (qualidade estética) → use `capricho` (evolução desta)
- Validação de ideia de negócio → use `triagem-ideias` / `validacao-ideia-24h`

## Conceito central

- **Depth Tree**: task → split recursivo N layers; leaves são único lugar de trabalho real. `tree 2-3` = feature solo, `4-5` = subsistema, `6-7` = projeto inteiro orquestrado.
- **Gates ledger** `.unlazy/<scope>/GATES.md` (solo) ou `.unlazy/<scope>/gates/leaf-*.md` (orquestrado): cada gate = checkbox + outcome observável + `CHECK:` (comando) + `EXPECT:` (substring ou /regex/) + `EVIDENCE:` (prova). `[x]` com `pending` = UNMET.
- **ABANDON honesto**: `ABANDON: <id> <reason>` → sai com `HANDOFF REQUIRED` (exit 1), nunca silent narrowing.

## Modos

| Modo | Tree | Como roda | Ledger |
|------|------|-----------|--------|
| **Solo** | 1-3 | uma sessão, sequencial | `GATES.md` único |
| **Orquestrado** | 4+ | leaves em subagentes paralelos, parent reverifica | `PLAN.md` + `gates/leaf-*.md` + `dispatch.json` |

## Workflow (solo resumido)

1. **Escreva gates antes do código** — 1 outcome observável por gate, com CHECK/EXPECT. Lint: `node <skill>/scripts/gate-lint.mjs GATES.md` (veja `references/gates.md`).
2. **Aprove explicitamente** — parse sem executar: `node <skill>/scripts/gate-check.mjs --status GATES.md` → leia cada CHECK/EXPECT/CWD/shell/PATH → ` --approve GATES.md`.
3. **Trabalhe cada leaf em 4 passes**: implemente completo → releia como expert → caçe bugs (corretude, integração, portabilidade, performance) → polish até passar limpo.
4. **Rode checker**: `node <skill>/scripts/gate-check.mjs GATES.md` — só marca `[x]` se exit 0 + EXPECT match; registra shell, CWD, exit, fingerprint.
5. **Re-verify antes de reportar**: rode de novo; se algum gate UNMET ou ABANDON, não reporte done. Reporte ledger completo com counts.

Orquestrado: veja `references/orchestration.md` — dispatch waves, lease, parent `--reverify`, Stop hook opcional.

## Gates que falham honestamente

- Use token de sucesso decisivo, exija exit 0 + EXPECT.
- Teste negativo contra controle positivo antes de confiar em ausência.
- Prefira scripts Node portáveis; não assuma `grep` no Windows.
- Ledger malformado, zero gates, id duplicado = erro, não done.

## Instalação (oficial)

```bash
npx skills add Leonxlnx/unlazy          # oficial MIT
# ou clone manual:
git clone --depth 1 https://github.com/Leonxlnx/unlazy ~/.claude/skills/unlazy
```

Esta skill ECC é um wrapper adaptado; scripts oficiais ficam em `<skill>/scripts/` quando instalada. Sem eles, use o ledger manualmente.

## Referências

Attribution

majinmagrosmajinmagros
View sourceMore from majinmagros →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

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

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

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 →