Thanks, I’ll keep following them for a while and try to get the hang of it. While the integration is similar to VW ID and WeConnect, there’s clearly a bit of a learning curve for me 
In case I find something that even after a while feels like it could be more clear with some changes, I head to GitHub to raise an issue or pull request… 
So I got around to testing you suggested setup for setting the maximum charge perccentage. I turned off the battery preservation feature and removed the home location. Now it works as expected by calling the service to set the max charge limit. And when the car starts driving, I set the charge limit back to 80% to restore battery preservation behavior. So any other limit is now one time only.
To get the unlock behavior back, I use the Enyaq device tracker to see if it is home. Then, when the car is plugged in, an automation sets the “auto plug unlock” feature on. That works. If the car is not plugged in, the switch turns back off, but not when it is plugged in. And finally when the car leaves home, I turn “auto plug unlock” back off. So basically I got back all I wanted, but just in a different way.
Thank you very much for pointing me the right way, it could have taken ages to figure out myself.
Glad it worked out!
Hello.
Anyone have make visualization of open door/windows/trunk etc of Karoq?
How its looks?
Can you share the images?
If there are others who have been suffering with the rate limiting problems, wanted to share my experience. I had the poll frequency at 60s (or was it even 30s) for some time and it worked fine, but few weeks ago every afternoon at the same time the inegration started reporting unavailable and came back alive at UTC 00:00. After increasing the poll frequency (having read about the problems both here and with VW), it was again stable for some days with 120s, but now it changed again a few days ago. Now the unavailability comes at 3PM UTC and as expected lasts until 12PM UTC. So it seem I have to increase the poll frequence again to at least 180s.
Why I wanted to share this is that it may not be enough to give any exact “safe” polling frequency. It would be nice to know how it works at the back end, why some clients are limited (seemingly) more than others. We do not have anything else polling the API except this integration, I assume the VW integration is working with a completely different back-end (although it could explain something if VAG was harmoninzing their e-services and rate limiting ip’s with too frequent connections)…
edit: I changed the poll frequency to 180s around mid day. Still, at exactly 3PM UTC the integration stopped working. Strange… 
from time to time, the connection is down :((
Now, when i try add a new integration, i get in log:
2023-11-03 19:43:01.338 DEBUG (MainThread) [custom_components.skodaconnect.config_flow] Creating connection to Skoda Connect
2023-11-03 19:43:03.363 ERROR (MainThread) [skodaconnect.connection] Token exchange failed. Request status: 429
2023-11-03 19:43:03.363 ERROR (MainThread) [custom_components.skodaconnect.config_flow] Login failed with error: Token exchange failed. Request status: 429
On website i can connect of course, same in oryginal app from skoda.
Same here since this PM. Also the 429 error.
I hope tomorrow i can connect, will the reduce the polling frequency.
Now the Skoda connect website has an internal error too, and the apps cannot login either. Looks like the vw group identity server is down.
The Skoda Connects Website [1] works, but the Integration can’t connect since fews days:
Could not login to Skoda Connect, please check your credentials and verify that the service is working
It seems the integration needs to be adjusted because the VW group changed something in the authentication:
Fix is ready… don’t ask me how they did it but it sure made my day
Quite a lot has changed under the hood last 6 months.
It seems there’s plenty of changed URLs when tracing the app and the login/token handling now has support for PKCE challenge/response which the latest MySkoda app uses.
Fortunately this time it was enough to bring the mimicked clients HTTP headers up to speed with what the recent app version uses.
It’s a ticking bomb though since there’s plenty of code that needs to be adapted sooner or later in order to make sure the integration continue working in the long run.
There’s also positive changes in the API, web sockets for instance which can greatly increased responsiveness and prevent the recent HTTP 429 errors. I’m not sure though if we can implement it in Home Assistant the way integrations work today.
thanks for your efforts
And I’m quite optimistic for the future 
After updating to 2023.12 the integration fails:
ERROR (MainThread) [skodaconnect.connection] Received unhandled error indicating client-side problem.
Try open your Skoda app and accept New condition. I have this without upgrade my ha to last
I had the same problem. Opening the app, accepting the new Ts&Cs and reloading the integration fixed it.
Thank you.
Having had this issue many times before, I knew to go to the app and do the same. Except for a couple of days this didn’t work nor prompt me.
It wasn’t until I went the Skoda Connect Website and logged in there - then I got the prompt to accept the T&Cs
I’ve got a possibly stupid question: what units does my Skoda (an elderly Kodiaq) or the Skoda Connect integration report in, and is there any possibility it might be set incorrectly?
I had an issue with the integration not connecting for a while (was a combo of new T&Cs and also the login issue – I hadn’t updated the integration in a while) and ended up deleting it/reinstalling.
Since then it works just fine but all of the average fuel consumption sensors (from the ‘last trip’ to the ‘long term’) are reporting ~2000mpg. I’m confident that my Kodiaq isn’t capable of travelling the circumference of the earth on a single tankful of fuel, so something must be wrong!
Is this a ‘new data, averages broken’ issue, or is it being reporting in meters per litre or something and being slapped with miles-per-gallon labelling by mistake, do you think?
EDITED TO ADD: in the MySkoda app, it reports a rather less pleasing but more reasonable 32mpg for my most recent trip, so at least somewhere, the data looks correct.
FURTHER EDIT: The integration has a configuration option for unit conversion. If I leave the units ‘unconverted’ then I get realistic km/l consumption figures which convert exactly to the mpg values shown in the MySkoda app. If I select the ‘imperial units’ conversion, I get the crazy numbers above. Is this a bug in the integration, a bug in Home Assistant’s unit conversion or some special case I’ve lucked into?
Any ideas why the odometer is saying I have 1400 miles on the clock, when there’s 9000? The Skoda app shows it correctly, but this one has it wrong. Thanks 
Hi there
I have a problem with my Skoda Kodiaq connected to HA via Myškoda app.
The entities are visible and workable, but after one day, or one ride they become unavailable.
I a checked the Myškoda app, there was the same, the car card was gone, and it say: missing data.
If i log in once again in Myškoda, the car is added, and the HA intergrations works well. But still only for one ride
Does anyone have a similar problem?
Maybe i need to accept he T&Cs, but i cant find, where is it on website
Any ideas?