Local iPhone Bluetooth presence and estimated rooms for Home Assistant. Install with HACS, enroll with Presence Pair, and keep presence data in your home.
Record exact app build, integration/receiver version, iPhone/iOS, adapter/driver and installation mode. No passwords, QR URLs, identity keys or packet captures belong in a public report. Software CI is not a hardware certification.
Start with the public-app smoke test. A pass is evidence for that recorded setup only, not a pass for every row below.
| Case | Expected result |
|---|---|
| Fresh, neither side paired | iOS consent if requested; protected ACK, HA association and phone completion |
| Valid bond on both sides | Reuse without unnecessary removal or repeated popup |
| Only iPhone remembers | Bounded recovery or actionable preparation; never false success |
| Only receiver remembers | Exact-target repair when available; no unrelated unpairing |
| Cancel, decline or timeout | Attempt stops within its budget; no partial success shown |
| Weak signal before connecting | Guidance to approach, not a claim that pairing completed |
| Move away during exchange | Clear interruption and bounded cleanup; new attempt remains usable |
| HA/MQTT interruption | No false completion; state recovers and retry remains possible |
| Explicit remove, then HA restart | Server removal verified; iPhone preparation survives restart |
| Second phone/person | No reassignment or deletion of the first phone |
| Passive tracking, screen locked | Fresh timestamps; stale reception never proves absence |
| Settings > People assignment, removal and other owner | Live panel status; no internal writes, lost GPS/Wi-Fi selection or forced re-pairing |
| Live signal with native HTTPS / trusted proxy | Correct public-certificate pin when available; no storage lookup or TLS bypass |
| Two receivers, away and return | Explainable estimates; no precise-distance promise |
For a clean-start test, explicitly remove only the selected phone in HA and wait for verified server removal, then forget that receiver in iPhone Settings if it is listed. This is a test preparation step, not a requirement for normal retries.
Use the normal public HA panel and actual app, not a diagnostic harness or simulated preview. Show app version, code creation (keep the recording private), scan or same-device launch, proximity guidance, any iOS consent, successful completion on the phone and HA, and a fresh presence/signal observation. The video must show the exact binary selected for review. A failed attempt is useful diagnostics, but is not evidence for a successful release gate.
Use the repository’s Hardware compatibility issue template for sanitized results. State exactly which rows passed, failed or were not tested. Linux/ARM and other adapter/iPhone combinations remain experimental until independently exercised.