Files
rickandCursor 1c2f8b6c07 Initial commit: Battery Lifetime integration v0.1.0
A Home Assistant custom integration that turns every numeric percent
battery sensor into an actionable predicted-replacement-by datetime,
with auto-detection of replacements, lithium/alkaline chemistry
profiles, a confidence ladder, cold-start backfill from long-term
statistics, and a forward-prediction service. HACS-installable.

Implements the add-battery-lifetime OpenSpec change:
- 56/62 tasks done (10.2-10.7 are manual-validation gates)
- 18 spec requirements across 5 capabilities
- 107 unit tests passing under pytest-homeassistant-custom-component

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-05-08 18:53:51 +02:00

23 lines
1.2 KiB
Markdown

# Battery Lifetime
A Home Assistant integration that turns raw battery percentage sensors into actionable
predicted-replacement-date sensors. Built for the operator who wants to answer
"which batteries do I need to replace this month?" or "which batteries should I
proactively replace before leaving the cottage for four months?".
For every Home Assistant entity with `device_class: battery` and unit `%`, the
integration creates a companion `*_replace_by` datetime sensor, a confidence
sensor, an observed drain rate sensor, and a small set of switches/buttons that
let you tell it about the chemistry, mark a battery as replaced manually, or
opt the battery out of tracking.
Replacement events are auto-detected from the source sensor's value (a
`<80% → ≥100%` jump within the last 30 days), with glitch and stale-prior
protections, and emit a Home Assistant event your automations can listen for.
A `battery_lifetime.predict_at` service forward-simulates every tracked
battery to a future date so dashboards can answer the cottage-departure question.
See `README.md` for the full feature list, profile defaults, and the documented
non-goals (rechargeables, voltage-only and categorical battery sensors,
multi-cell awareness).