hey @felixrieseberg ! been hacking on the desktop buddy firmware, got it showing AskUserQuestion on the device with the buttons mapped to options. picking works. looks great.
but the bridge only accepts once/deny back, so the choice never makes it home. Claude just sees "no preference" no matter what you pick.
any chance you'd be open to extending the bridge to accept a choice from the device?
@grooveygr appreciate that. this is what i was aiming for, tired of reading setups i can't actually reproduce. all these posts about 50 parallel cc instances shipping nothing but screenshots.
hope this brings you back to our contrib charts, miss seeing you up there!
most posts about "running parallel AI agents" talk about the number of agents.
nobody talks about what happens when 3 of them finish at the same time and you can't remember what any of them were doing.
i went through the whole progression. manual branch switching, cloning the repo 3 times, gitbutler. nothing stuck until i landed on git worktrees.
the setup is embarrassingly simple. wrote it up with real config and real rough edges.
היופי זה להצליח לקחת את ה20% ולשים את האצבע על מה הם מכילים, ולעבוד עם הכלים בשביל לטייב סקיל/דוק/טסט בשביל שפעם הבאה זה יהיה טוב יותר. באקסטרים אפשר אפילו למחוק את התיקונים הידניים לקוד אחרי החידוד שאתה עושה ולחזור על התהליך עד שאתה לא ניקית כלום ידנית.
זה אקספוננט ומהר מאוד לא כותבים כלום לבד.
Every time should be a little bit better
@n2parko@cursor_ai Awesome changes!q
Do you plan supporting multi model, review and judge on large monorepos soon? Could really help large teams move faster!
Every site published on Lovable is now scanned by @GuardioSecurity for phishing, scams, and impersonation.
As AI-powered tools democratize digital creation, we ensure these new tools are not used to cause harm.
This is what responsible platform-building looks like.