agent-token-efficiency

No explicit credential-stealing or exfiltration logic is shown in the provided instructions, and the content appears to be setup guidance. However, it contains a high-impact supply-chain execution pattern (`irm ... | iex` on a network-fetched script) with no visible integrity verification, and it instructs installation/activation of multiple third-party components (npm/pip/npx/plugin marketplace) with global scope and an optional local proxy/hook that can mediate agent API traffic. Treat as potentially dangerous unless dependencies are pinned and verified, and consider sandboxing/least-privilege for installation steps.