Thank you all for being part of the launch yesterday, all the people I got to meet, all the experiences throughout the day, the pitches, the competition... that's all part of it, and I couldn't be happier
I also got some really nice insights about Toone from the community. A lot of people showing genuine interest in the tool and in what comes next.
I don't want to spoil too much, but we have a mobile application coming for iOS that will allow you to control Toone, dispatch and watch routines, and talk to an agent as if it were running on your machine. Apple just gotta approve.
We've also already implemented a sharing functionality for routines, allowing people to make their expertise and workflows available to the public.
@mujtabaidrees94 Hi Mujtaba, thanks for the comment and kind words. Yes totally, one is able to fork or simply clone a routine (updates on them by the author gets them updated). Thanks for the question, on point!
Congrats on the launch, Matheus! What I like most is that a node can be either an agent's chat thread or a plain step that runs the same way every time without AI. That's the right split: let the model handle what needs judgment, and keep everything else predictable. Being able to pause a routine and continue later is a nice touch too.
@matheus_paranhos1 you are crushing it again! congrats on going live! pause and resume mid-run is such a cool feature, n8n flows breaking halfway is my daily pain 😅 does it work with routines that pull from Calendar too?
@kate_ramakaieva Haha thank you Kate! Answering your question: Yes! it flows really well with calendar, agents have their own single events and routines are reoccurring by default. Thanks for showing up!
i’m all about stealing clever routines from folks who really know their stuff. might be a bit before i feel like i've got any worth sharing back, but hey, that’s future me’s problem.
@harini_mukesh Yeah right!? There's a lot AI needs to be polished and having routines straight from experts makes a total difference in the end result! Thanks for the comment 🙌
i'm all about trying out different routines, but the ability to share them with others is kinda cool. i might get inspired by someone's workflow and actually stick to it for once, fingers crossed.
Nice launch, @matheus_paranhos1 Writing a job in plain language is a low barrier to start, but I'd guess the hard part is getting from a first attempt to a routine you can rely on. How do new users learn what makes a good instruction? Are you collecting example routines in the docs, or is that mostly learned inside the app for now?
@roopreddy Yeah, agents can access documentation. The whole premise is that an agent can execute and troubleshoot everything and know exactly what's going on. The approach is balanced so that it won't go for advanced patterns that would consume a lot of tokens or be excessively rigorous unless the user points out his needs and the agents find that necessary. Thanks for your question!