I often use the template editor to bring an idea to life in a data set, and then I iterate my template logic as I refine it. I’m pretty sure that this is normal for us Jinja Junkies (™). But I often recreate it without removing the previous work just in case I completely screw up the previous work. Well today I went a little to far and got this…
That’s why you should learn how to remove your whitespace which counts against the character limit.
Don’t mistake this for a limit in just the template editor. This is a limit for all templates.
Ya that was likely it, I had a lot of | to_json(pretty_print=true) going on so that I could see the data sets to determine which one made the most sense.
Per script or per active run? Like if I had an automation that called a script (or 2) which also called other helper scripts along the way to return values while updating template entities as they went (theoretically speaking of course 
), is that the limit of the sum of all of those or each script/file rendering before returning values?
Per single template.
Anywhere a template renders
You can have something in memory all you want but the minute you try to render it…
For me the biggest hazard is the output of the LLM prompt template on a conversation prompt or aitask prompt - I ran into this early building Friday.
That happened to me, too. I copied everything to a text file, saved the file on my NAS and then deleted everything in the template editor. I thought I might come back to it, but I haven’t yet.
I use Notepad++ as a desktop text editor.
