Anthropic fixed the flaws - but the AI-enabled attack surfaces remain Security vulnerabilities in Claude Code could have allowed attackers to remotely execute code on users' machines and steal API ...