Analysis updated 2026-05-18
Get a computed weather forecast and comfort ratings in Home Assistant from a WeatherFlow Tempest station.
Track rain totals, lightning strike counts, and fog or snow probability as Home Assistant sensors.
Replace an archived weather integration with a maintained fork that includes bug fixes and dependency updates.
| vatoe/hass-weatherflow2mqtt | 0xallam/my-recipe | 0xhassaan/nn-from-scratch | |
|---|---|---|---|
| Stars | 0 | — | 0 |
| Language | Python | Python | Python |
| Last pushed | — | 2022-11-22 | — |
| Maintenance | — | Dormant | — |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 3/5 | 2/5 | 4/5 |
| Audience | general | general | developer |
Figures from each repo's GitHub metadata at analysis time.
Needs a running MQTT server and Docker on your network.
This project is a Home Assistant add-on that connects a WeatherFlow weather station, either the Tempest device or the older AIR and SKY devices, to Home Assistant through MQTT. It listens for weather data broadcast by the WeatherFlow Hub over your local network and publishes that data to a MQTT server, where Home Assistant can pick it up and turn it into sensors automatically. This is a fork of an earlier add-on by Bjarne Riis, which was archived after Home Assistant added its own official WeatherFlow integration. That built-in integration only exposes the raw numbers the weather station sends. This fork's purpose is to keep offering something the built-in integration does not: computed sensors built on top of the raw data, including a pressure based weather forecast, comfort level ratings for temperature and UV, chances of fog or snow, cloud base height, sea level pressure, rain totals, and counts of lightning strikes over recent time windows. Beyond keeping the original add-on alive, this fork fixes a bug where comparing a new sensor reading against a missing stored value would crash the update process for every sensor at once, not just the one causing the problem. It also isolates failures so a problem with one sensor no longer affects the others, and it protects the underlying database queries by rejecting invalid numeric readings instead of writing bad values or crashing outright. A handful of aging dependencies have also been updated, and the project's base software image was refreshed too, since the older one had stopped receiving updates. To use this, you need a MQTT server running somewhere on your network, which most Home Assistant Supervised installations can add easily, and you run this add-on as a prebuilt Docker container. Once running, it watches your WeatherFlow Hub's local network broadcasts and automatically creates matching sensors in Home Assistant using MQTT Discovery. If you already get the raw numbers you need from Home Assistant's built in integration, you likely do not need this. But if you want the added weather analysis layer describing conditions in more useful terms, this add-on still provides that, along with ongoing maintenance the original no longer receives.
A Home Assistant add-on that connects WeatherFlow weather stations to MQTT and adds computed sensors, like storm forecasts and comfort levels, that Home Assistant's built-in integration lacks.
Mainly Python. The stack also includes Python, Docker, MQTT.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.