I’ve been having stability issues with Weatherflow to MQTT lately (restart HA resolves the issue) so I’ve been trying the new Forecast integration. I don’t like that it depends on the cloud and it much slower to update which is expected.
The plain Weatherlow official integration give me a UDP discovery error, is this because I’m still using the MQTT integration through HACS? What is the best method to reset and start over with the latest local only integration?
Just joining the party, as I installed a Tempest last night. My initial observation was lightning strikes. Using the integrated weatherflow, I see this in HA:
And I see this in the Tempest online view:
I agree with Tempest. No lightning here last night or today. Yet, HA is reporting counts continually…
Is this solved in the HACS version or other way?
Weatherflow filter out lightning data to account for the way the lightning sensor in the Tempest works - what gets reported via the Tempest app/online shows the filtered results
WeatherFlow appears to support 100 RESTApi hits / minute - once the sensor components get added into HA I’ll be moving from a 15min update cycle to a 5second update cycle.
There is def stability with the integration right now … and I just found another condition - I’m on handling which will throw an error in the logs but not really mess with stuff.
{"status":{"status_code":1,"status_message":"INTERNAL_ERROR - null"}}
Again → if you need stability and more features go with hacs. If you want to be helpful and find all the bugs use core 
Does anyone know where forecast probability went? I used this reading quite often to send signals to the house if it was going to rain today?
I looked into buying a Tempest but then saw the local map of existing installations at for example here, and there are already hundreds if not thousands of devices installed in my area. It makes no sense to buy one if I can get enough data from the existing ones.
As a thought experiment on using ChatGPT to do most of the programming for me, I just spent a few hours prompting and driving ChatGPT4o to create a python script to obtain the data for any weather station in the world, without requiring a login or token. It gets the same data as other approaches do.
It formats it in json then uploads to an MQTT server for use in HA. The script handles parameters and config files, interactively lets you select a station if there are several to choose from, and a few other things.
I then used ChatGPT to create a myriad of MQTT sensors and templates and a weather dashboard. I also wanted a windrose, so the script generates windrose data as well for use in Grafana. The grafana graphic is then iframed into my weather dashboard.
Obviously the main problem with this approach is that it depends on a website remaining unchanged. Any major revisions will likely break the script. But for now, it works, generates the same 20 or so sensors, and works fine.
@SpinCharm Can you share the code you used to do this? My neighbor has one so it seems silly to buy one as well…
Is it possible to have a working local configuration of the integrations, while completely blocking internet access to the Tempest (for example, through my router)?
Not sure you need to block it at your router. I believe Tempest is good about honoring ‘not sharing publicly’. But, you certainly could block it if you want.
And, yes. There are two integrations. WeatherFlow and WeatherflowCloud. Just don’t use the second one.