Platformizing the Body Area Network: An Academic Scrutiny of DOORM's Inner/Outer Environmental Domain Visualization Platform — Genmount WorldModel-OS as Robot Brain

This preprint is a public-material system scrutiny of DOORM AI's Personal Information Tower page (https://personalinformationtower.com/zh) presenting an embodied body-area-network intelligence · inner/outer environmental domain visualization platform. It operationalizes platformized WBAN (dimensions D1–D5), abstracts scenario replays as a replayable task graph, and interprets the stack through Genmount WorldModel-OS as a robot-brain / world-model runtime (body domain → environment domain → world model → actuators → audit chain). The paper discusses reconfigured engineering bottlenecks, actionability vs over-medicalization, and non-clinical audit verifiability, and freezes testable propositions plus an evaluation roadmap. No measured closed-loop latency, false-alarm rates, AUC, or retention numbers are claimed. Chinese and English full texts are deposited with matching section numbering (Sections 1–7 and Appendices A–D). A small helper (bodynet_ref.py) is included with no filled result tables. Version DOI: https://doi.org/10.5281/zenodo.23413279 (do not invent Concept DOI).