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

Goal

ASecurity

Use when the user runs /goal to set a durable objective with an acceptance check for a long-running session. Triggers on "meta", "objetivo", "goal", "critério de aceite". Pins the objective, keeps working toward it, and does NOT declare completion until the acceptance criteria actually pass (real test runs, not vibes).

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

Works with

api

Security Analysis

A100/100

Scanned 9/19/2026

Install to Claude Code

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

Installs into .claude/skills of the current project.

Are you the author of Goal?

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

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

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

Download Zip
Files
SKILL.md
---
name: goal
description: Use when the user runs /goal to set a durable objective with an acceptance check for a long-running session. Triggers on "meta", "objetivo", "goal", "critério de aceite". Pins the objective, keeps working toward it, and does NOT declare completion until the acceptance criteria actually pass (real test runs, not vibes).
---

# Skill: Goal — objetivo durável com auditoria de conclusão

Fixa um objetivo + critério de aceite para guiar a sessão inteira. Só encerra
após PROVAR o aceite com checros reais.

## Quando usar (gatilhos)

- "/goal"
- "Define a meta dessa sessão"
- "Só termina quando o teste passar"
- "Fixa o objetivo e o critério de aceite"
- "Não declara done sem prova"

## Exemplo

```markdown
## Goal: checkout sem erro 500
Proof: `npm test checkout` verde + pedido cair no banco
Stop: se precisar mudar schema, me pergunta antes
```

## 1. Capture o goal (objectivo específico)

Antes de começar, registre em disco (`.opencode/goals.md`, ou use o task list):

| Campo | O que escrever | Por que importa |
|---|---|---|
| **Outcome** | Uma mudança observável | Evita projeto de limpeza infinito |
| **Escopo** | Diretórios/serviços/pacotes restritos | Mantém o raio legível |
| **Constraints** | APIs, schemas, arquivos que NÃO podem mudar | Protege compatibilidade |
| **Proof** | Testes/checks/renderização EXATOS | Dá alvo real ao goal |
| **Stop rule** | Condições que exigem decisão sua | Não vira edição sem controle |

## 2. Trabalhe rumo ao goal

- A cada bloco (~10 model calls) sem progresso aparente, releia o goal e
  reporte o estado/continue — não deixe a sessão derivar.
- Não declare vitória com critério vago: "goals, non-goals, constraints, risks,
  validation criteria must be clear" antes de marcar como done.

## 3. Auditoria de conclusão

- Para fechar o goal, rode as checros/ORÁCULO exato do campo **Proof** (ex.:
  rodar o teste com os 6 checks passando).
- Se o oráculo não passou, NÃO está done — corriga até passar ou REPORTE o bloqueio.
- Ao final, reporte: o que passou, o que mudou no caminho, e o link do proof.

## 4. Salvaguaras

- Um objective mal escrito (sem oráculo) permite "positivamente inexistente".
  Se o goal vier vago, refine com /grilling ou pergunte até ficar mensurável.

## 5. Template provable + verifier loop (leva YouTube rodada 5)

Template canônico do goal:

```markdown
## Goal: <one measurable end state>
Prove it by: <oráculo exato>
Constraints: <o que NÃO pode mudar>
Stop after: <N turns sem progresso → parar e reportar>
```

- **Verifier separado** — um modelo barato (ex.: Haiku) dá veredito a cada
  turno/bloco: goal avançou, estagnou ou derivou. Veredito `stalled 2x` = parar,
  não insistir.
- **Auto-mode explícito** — declare se o goal roda autônomo (sem aprovação por
  passo) ou com gates. Autônomo sem stop rule é loop infinito pago.
- **Fan-out com prova por branch** — goal paralelo (N agentes) exige proof
  independente por branch antes do merge; branch sem proof não integra.

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 →