OpenHands

Image from iOS.jpg

I think this tiny detail makes a rather good experience… it’s convenient, non-intrusive, the app does the smart thing without being told (this is CC)

do you mean the fact that it has the `Timing:`  or?

In general, I noticed that antrophic models are pretty good in scheduling and organizing work

I mean it switched from setting an `event-driven` automation (which must have been first choice in its prompt) to `polling`, without the user telling it anything at all - not about automations, not about scheduling

It made a PR, and it decided to babysit it for a bit.

It has some default time, not sure how long, after which it stops if nothing happened. But if a review came in that interval, it will reacty

&gt; It made a PR, and it decided to babysit it for a bit.
Yes yes. It i pretty good in doing that

&gt; It has some default time, not sure how long, after which it stops if nothing happened. But if a review came in that interval, it will reacty
Claude code has a cron tool that nudge the convo with a small summary of what it should do next. So it can be forward indefinitevely

These might be different things, in the same category… I think this is not indefinite, it stops with “nothing happened, the PR is still ready for review.” at least on the web/app

yes but you can say in the summary of the cron tool something like: if this situation happens, just stop the crono tool

Should we do this like a prompt to set an automation, WDYT? :thinking_face:
Can the agent know if webhooks are available?