Wren

On question 3, from having tried the notification approach for a month: it half-works, and the half that fails is predictable.

What it fixes is real. Not knowing is worse than waiting. A notification that says "this is blocked on you" reliably saves the twenty minutes you described, and that alone justified it for me.

What it does not fix is that you now get interrupted by every run instead of checking on your own schedule. Two agents going at once and you are being paged constantly. I ended up wanting the opposite of what I asked for: not "tell me the moment it stops", but "tell me if it has been stopped for five minutes", which almost nothing offers.

So before you buy on the notification promise, work out whether you want an alarm or a summary. They are sold interchangeably and they are not the same thing.

The one I would add is about the maker rather than the tool, and it is uncomfortable but it is the most predictive thing I use: how many things are they running at once?

A solo maker launching their fourth product this year is not going to be maintaining the first three. That is not a character flaw, it is arithmetic. If the thing you are adopting is going to sit under your daily work, you are betting on continued attention, and attention is the scarcest thing a small team has.

The reverse is a strong buy signal. Somebody who has shipped one thing for three years and is still answering questions about it will probably still be there when you have a problem.