ARCHITECTURE
- Mobile AR Digital Twin MVP skeleton for cross-vendor assets (PV, inverter, wind, storage)
- Core: Python-based digital twin engine with TelemetrySample, AssetModel, TwinEngine
- API: FastAPI scaffold for telemetry ingestion and health endpoints
- Offline caching layer (intentional MVP kept in-memory for simplicity; ready to swap to SQLite)
- Data contracts and adapters: TelemetrySample, AnomalySignal, Alert, GovernanceLog (minimal stubs)

TECH STACK
- Language: Python 3.9+
- Core: ar_grid_tutor_mobile_ar_digital_twin_for.core
- API: api.app (FastAPI)
- Tests: tests/ with pytest
- Packaging: pyproject.toml with setuptools

TESTING & RUNNING
- Run unit tests: ./test.sh (requires pytest and build tooling)
- Build package: python3 -m build
- Run API locally (dev): uvicorn api.app:app --reload --port 8000

CONTRIBUTING RULES
- Use the provided test.sh to verify local changes before publishing
- Do not modify READY_TO_PUBLISH unless you completed a publishable milestone
- AGENTS.md is the canonical onboarding doc for agents; do not delete

NOTE
- This is an MVP skeleton designed for extension; future work will flesh out offline storage, adapters, and governance modules.