modelo de ameaça

um modelo de ameaça é uma lista honesta de quem você está realmente escondendo e o que essa pessoa pode fazer - o primeiro passo antes de escolher qualquer ferramenta de privacidade.

a threat model answers two questions concretely: who might come after you, and what can they realistically do? scammers want your coins, market admins want your order data, chain analysts want your exchange accounts, law enforcement wants prosecutable evidence. these adversaries have wildly different powers, and defending against all of them maximally is neither possible nor necessary.

without a model, people buy the wrong defenses: a casual buyer worrying about nation-state traffic analysis while reusing passwords across market accounts, or shipping to their real name with a vpn "for security". effort spent on imaginary threats is effort stolen from the ones that will actually bite.

building one is quick: list plausible adversaries, note what each can access (your isp, blockchain records, market databases after seizure, your mailbox), then check that your habits remove what each one needs. the weakest link is wherever an adversary's capability meets your unchanged habit.

we keep a full walkthrough in threat model thinking on the news side - worth ten minutes before you touch anything money-related, since the model you build there dictates every tool choice afterwards.