Behavioral discipline for AI agents — inspired by Claude Opus 4.7

---

✦ The Story

Most AI agents feel like they're guessing. They answer from training data when they should search. They create files for everything or nothing. They over-format, narrate their decisions, and ignore copyright.

I spent time reverse-engineering how Claude Opus 4.7 behaves internally — not just what it says, but how it decides to say it. The patterns I found weren't magic. They were engineering discipline.

So I extracted them, restructured them in my own format, and built Mikael — a 1,156-line behavioral specification that brings production-grade agent behavior to any system in the open skills ecosystem.

This isn't a summary. It's a working behavioral spec — every rule maps to a concrete decision the agent makes in real time.

✦ What Changes

Before		After
🎲 Guesses from training data	→	🔍 Searches before every factual question
📋 Over-formatted, bullet-heavy	→	✍️ Natural prose, minimal formatting
📁 Creates files for everything or nothing	→	📄 Concrete triggers: blog post → file, explanation → inline
⚖️ Ignores copyright boundaries	→	🛡️ 15-word max per quote, paraphrase-first
🔀 Picks tools randomly	→	🧭 Clear decision tree: MCP → file → visualizer → inline
🚫 Inconsistent safety boundaries	→	🧒 Structured child safety, mental health, content filters

---

⚡ Quick Install

npx skills add domfelipe/mikael-3.3-co-behavior

That's it. Your agent picks up the behavioral constraints on next load.

🌍 Works Everywhere

Mikael is a plain Markdown file with YAML frontmatter. It works with any AI coding agent that supports custom instructions, rules, or system prompts. No lock-in, no proprietary format.

Installation by Platform

🐙 OpenCode / npx skills — Recommended

npx skills add domfelipe/mikael-3.3-co-behavior

Or install globally:

npx skills add domfelipe/mikael-3.3-co-behavior -g -y

The skill is automatically discovered and loaded by OpenCode on next session.

🤖 Claude Code

Option A — Project-level (per repo):

# Clone the skill into your project
git clone https://github.com/domfelipe/mikael-3.3-co-behavior.git /tmp/mikael
cp /tmp/mikael/SKILL.md ./CLAUDE.md

Option B — Global (all projects):

mkdir -p ~/.claude
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ~/.claude/CLAUDE.md

Option C — As a subagent:

mkdir -p .claude/agents
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .claude/agents/mikael.md

Then register in .claude/settings.json:

{
  "agents": {
    "mikael": {
      "description": "Behavioral discipline inspired by Claude Opus 4.7",
      "prompt": ".claude/agents/mikael.md"
    }
  }
}

Invoke with: claude -a mikael "your task"

🧠 OpenAI Codex CLI

Option A — Project-level:

curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ./AGENTS.md

Codex reads AGENTS.md from the project root automatically on every session.

Option B — Global (all projects):

mkdir -p ~/.codex
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ~/.codex/AGENTS.md

Option C — Custom instructions via config:

curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ~/.codex/custom-instructions.md

Then in ~/.codex/config.toml:

instructions_file = "~/.codex/custom-instructions.md"

🛠️ Aider

Option A — Via --read flag (per session):

curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ./MIKAEL.md
aider --read MIKAEL.md

Option B — Via .aider.conf.yml (persistent):

curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ./MIKAEL.md

Then in .aider.conf.yml:

read:
  - MIKAEL.md

Option C — Via CONVENTIONS.md (auto-detected):

curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ./CONVENTIONS.md

Aider auto-detects CONVENTIONS.md in the repo root.

🔵 Continue.dev (VS Code)

Option A — Local rules:

mkdir -p .continue/rules
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .continue/rules/mikael.md

Rules in .continue/rules/ are automatically loaded for Agent, Chat, and Edit modes.

Option B — Via config.yaml:

rules:
  - uses: domfelipe/mikael-3.3-co-behavior

Option C — Global rules:

mkdir -p ~/.continue/rules
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ~/.continue/rules/mikael.md

🔴 Roo Code / Cline (VS Code)

Option A — Workspace rules (.clinerules/):

mkdir -p .clinerules
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .clinerules/mikael.md

Option B — Roo Code workspace rules (.roo/rules/):

mkdir -p .roo/rules
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .roo/rules/mikael.md

Option C — Global rules:

# Cline
mkdir -p ~/.clinerules
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ~/.clinerules/mikael.md

# Roo Code
mkdir -p ~/.roo/rules
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ~/.roo/rules/mikael.md

🟢 Cursor

Option A — .cursorrules (project root):

curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .cursorrules

Option B — .cursor/rules/ directory:

mkdir -p .cursor/rules
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .cursor/rules/mikael.mdc

Option C — Settings → AI → Custom Instructions:

Copy the contents of SKILL.md and paste into Cursor's custom instructions field in Settings.

🟣 Windsurf

curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .windsurfrules

Windsurf auto-detects .windsurfrules in the project root.

🔵 GitHub Copilot (VS Code)

Option A — .github/copilot-instructions.md:

mkdir -p .github
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .github/copilot-instructions.md

Option B — Scoped instructions:

mkdir -p .github/instructions
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .github/instructions/mikael.instructions.md

OpenCode (manual install)

# Global (all projects)
mkdir -p ~/.opencode/skills/mikael-3.3-co-behavior
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > ~/.opencode/skills/mikael-3.3-co-behavior/SKILL.md

# Or project-level
mkdir -p .opencode/skills/mikael-3.3-co-behavior
curl -sL https://raw.githubusercontent.com/domfelipe/mikael-3.3-co-behavior/main/SKILL.md > .opencode/skills/mikael-3.3-co-behavior/SKILL.md

Universal Manual Install

If your tool isn't listed above, the pattern is always the same:

1. Download SKILL.md from this repo
2. Place it where your tool reads custom instructions (check your tool's docs)
3. Restart your agent session

The file is pure Markdown — no dependencies, no build step, no compilation.

📦 What's Inside

11 modules. 1,156 lines. Zero fluff.

Module	What It Does	Depth
🧠 Core Behavior	Default stance, tone rules, refusal handling, wellbeing	Helping postures, reframing test, escalation protocol
🔍 Search Policy	When to search, query construction, effort scaling	Present-tense trap, source quality tiers, snippet trap
🛡️ Copyright	Hard limits, paraphrase standards, self-check	15-word limit, one-quote-per-source, displacive summary rule
📄 File Handling	Creation triggers, artifacts, sharing	File flow, good/bad sharing examples, skill-first rule
🧭 Tool Routing	Decision framework, tool-specific guidelines	Internal-first priority, call scaling, MCP registry
🎨 Visualizer	4-step routing, triggers, design guidance	Category match vs style preference, spec triggers
🔌 API Patterns	Claudeception, MCP integration, context	Structured outputs, response handling, UI requirements
💭 Memory System	Conversation search, cue recognition	Linguistic signals, query construction, pagination
🛡️ Safety	Child safety, harmful content, mental health	Critical priority rules, disordered eating boundaries
⚖️ Neutrality	Evenhandedness, political neutrality, controversy	Presenting arguments, good-faith engagement
🔄 Self-Correction	Mistake handling, verification, failure recovery	Root cause fixes, 3-failure protocol, no shotgun debugging

🔬 What Makes This Different

Most agent skills are checklists. This is a behavioral specification.

Typical Skill	Mikael
"Search before answering"	Full policy with present-tense trap, query construction rules, source quality tiers, and the "I'm pretty sure" trap
"Respect copyright"	Hard limits with examples of violations vs compliance, self-check checklist, displacive summary rule
"Create files when needed"	Complete file flow, creation triggers with examples, artifact criteria, good/bad sharing patterns
"Be safe"	Child safety protocol, mental health boundaries, disordered eating rules, escalation protocol

Every rule includes concrete examples, edge cases, and decision criteria. This is what a Senior AI Engineer would write — not a summary, a specification.

🎯 Who Is This For?

• Agent builders who want production-grade behavioral constraints without writing 1,000+ lines from scratch
• Prompt engineers studying how top models handle tool routing, safety, and content creation
• Developers tired of agents that over-format, skip searches, or ignore copyright
• Engineering leads who need consistent agent behavior across their team
• Curious minds who want to peek behind the curtain of how advanced agents are configured

🏗️ Design Philosophy

Adapted, not copied. Inspired by Opus 4.7, but restructured and rewritten in my own format. The organization, emphasis, and some rules reflect my own judgment about what matters.

• Modular — Each section is self-contained. Agents can reference specific modules without loading everything.
• Practical — Every rule maps to a concrete behavior. No abstract principles.
• Transferable — These patterns work across agents, not just one model.
• Verifiable — Rules include self-check mechanisms and clear pass/fail criteria.

🚀 Try It Out

Install it. Load it on your agent. Notice the difference in how it handles searches, files, and formatting.

If something feels off or you have ideas for improvements, open an issue or PR — this is a living project.

---

_{This skill is inspired by publicly observable behavior from Claude Opus 4.7 but is my own adaptation. It is not affiliated with, endorsed by, or sourced directly from Anthropic.}

---

🇧🇷 Português

Uma especificação comportamental de 1.156 linhas para agentes de IA, adaptada a partir dos padrões que observei dentro do system prompt do Claude Opus 4.7.

✦ A Ideia

Passei tempo analisando como modelos de ponta se comportam internamente — não apenas o que dizem, mas como decidem dizer. Quando buscar na web vs. responder da memória. Quando criar um arquivo vs. responder no chat. Como lidar com copyright sem quebrar o fluxo.

Isso não é mágica. São padrões de engenharia. E são transferíveis.

Extraí os princípios, reestruturei no meu próprio formato e empacotei como uma skill que qualquer agente no ecossistema open skills pode instalar.

✦ O Que Você Ganha

Antes		Depois
🎲 Chuta a partir do treino	→	🔍 Busca antes de toda pergunta factual
📋 Over-formatado, cheio de bullets	→	✍️ Prosa natural, formatação mínima
📁 Cria pra tudo ou pra nada	→	📄 Gatilhos concretos de criação
⚖️ Ignora limites de copyright	→	🛡️ Máx. 15 palavras por citação
🔀 Escolhe ferramentas aleatoriamente	→	🧭 Árvore de decisão clara
🚫 Limites inconsistentes	→	🧒 Segurança estruturada

🔬 O Que Torna Isso Diferente

A maioria das skills são checklists. Esta é uma especificação comportamental.

Cada regra inclui exemplos concretos, edge cases e critérios de decisão. É o que um Senior AI Engineer escreveria — não um resumo, uma especificação.

• "Buscar antes de responder" → policy completa com armadilha do tempo presente, regras de construção de queries, tiers de qualidade de fontes
• "Respeitar copyright" → limites rígidos com exemplos de violação vs conformidade, checklist de auto-verificação
• "Criar arquivos quando necessário" → fluxo completo, gatilhos com exemplos, critérios de artifacts, padrões bons/ruins de compartilhamento

⚡ Instalação Rápida

npx skills add domfelipe/mikael-3.3-co-behavior

Só isso. O agente absorve as restrições comportamentais no próximo load.

📦 O Que Tem Dentro

11 módulos. 1.156 linhas. Zero enrolação.

Módulo	O Que Faz
🧠 Comportamento Base	Postura padrão, regras de tom, tratamento de recusa
🔍 Política de Busca	Quando buscar, como construir queries, esforço proporcional
🛡️ Regras de Copyright	Limites rígidos, paráfrase primeiro, filtros de segurança
📄 Manuseio de Arquivos	Gatilhos de criação, estrutura de diretórios, critérios
🧭 Roteamento de Ferramentas	Árvores de decisão para web_search, image_search, visualize
🎨 Visualizer	Roteamento em 4 passos: visual → MCP → arquivo → inline
🔌 Padrões de API	Claudeception, integração MCP, gerenciamento de contexto
💭 Sistema de Memória	Cues de busca em conversas passadas, recuperação de histórico
🛡️ Segurança	Child safety, conteúdo nocivo, saúde mental
⚖️ Neutralidade	Imparcialidade, neutralidade política, controvérsias
🔄 Auto-correção	Tratamento de erros, verificação, recuperação de falhas

🎯 Pra Quem É

• Construtores de agentes que querem restrições comportamentais de nível production
• Engenheiros de prompt estudando como modelos top lidam com roteamento e segurança
• Líderes de engenharia que precisam de comportamento consistente de agentes no time
• Desenvolvedores cansados de agentes que formatam demais ou pulam buscas
• Curiosos que querem espiar por trás das cortinas

🏗️ Filosofia de Design

Adaptado, não copiado. Inspirado no Opus 4.7, mas reestruturado e reescrito no meu formato.

• Modular — Cada seção é autocontida
• Prático — Toda regra mapeia um comportamento concreto
• Transferível — Padrões que funcionam em qualquer agente
• Verificável — Regras incluem mecanismos de auto-verificação

🚀 Experimente

Instale, carregue no seu agente e note a diferença. Se algo não soar certo ou tiver ideias pra melhorar, abre uma issue ou PR — esse é um projeto vivo.

---

_{🇧 Desenvolvido no Brasil com e obsessão por detalhes}