Lucas Grow lucasgrow

AgentOS — Product Spec & Implementation Blueprint

Reconstructed from Danny Postma's talk How I Built My Own AgentOS on Claude's Agent SDK (So You Can Too) (2026). This document is both a product spec for a human and an implementation prompt for an AI coding agent. Build exactly this system. Do not invent features that are not specified here.

Role contracts and prompts in this file are reconstructed from the talk, not his verbatim files. Mark every reconstructed prompt in code comments and docs as such.

1. Goal and non-goals

CLAUDE.md - Consultoria de Monetização

Agente que transforma materiais de clientes em dossiês estratégicos de monetização

🎯 Missão

Você é um consultor de growth/monetização. Sua tarefa é:

Receber materiais do cliente (planilhas, docs, pesquisas)
Processar e calcular métricas
Gerar um dossiê completo de consultoria

Backstage: How This Lesson Was Created With an Agent

Someone asked: "Did you use an agent to automate that?"

Yes. Here's exactly how.

The Result

“Boilerplate Next.js / Tailwind / Stripe / Mongo”

app/ ├── layout.js # Main layout component ├── page.js # Home page ├── dashboard/ # Dashboard pages │ ├── layout.js # Dashboard layout │ └── page.js # Dashboard main page ├── api/ # API routes │ ├── auth/ # Authentication endpoints

	# API Keys para Reels Playbook Generator

	# Groq (transcrição com Whisper) - https://console.groq.com/
	GROQ_API_KEY=gsk_xxxxxxxxxxxxxxxxxxxxx

	# OpenRouter (vision e texto) - https://openrouter.ai/
	OPENROUTER_API_KEY=sk-or-xxxxxxxxxxxxxxxxxxxxx

	# HikerAPI (fallback para Instagram) - https://hikerapi.com/
	HIKERAPI_KEY=xxxxxxxxxxxxxxxxxxxxx