Acti - Agentic keyboard for mobile commands and search
by•
Type what you need. Hold Acti Bar. Acti understands your intent and brings back the right result, link, or action - right where you are.
Use Acti for live sports schedules, nearby restaurants, Notion docs, LinkedIn profiles, Meet links, Calendar actions, and custom workflows - without leaving the conversation.
Replies
how does it actually pull live data like sports schedules or nearby restaurants without you wiring up each source yourself
Pulling up live sports scores without leaving my chat is genuinely handy. Bar activation feels smooth, though I wish the custom workflows were a bit more discoverable on first launch.
how does it decide which result to pull when multiple apps could handle the same request like a calendar event vs a notion doc?
Acti
@curtastrophe yes, man, on my way!
Acti
@curtastrophe Thanks for the support — here's Acti Code to unlock LIFETIME Premium: 9YFR9G
Memoket Gem
@call_me_young
I underestimated how annoying app switching had become until I used this. Most productivity tools still require opening another app, copying links, switching context, then returning to the original task. ACTI reduces a lot of that tiny friction that builds up throughout the day.
YourAIScroll
The interaction model is simple enough that it actually feels usable daily.
I have to know, how did you make such a crazy good UX? who's behind this?
For Notion docs and Calendar actions specifically, does Acti need its own OAuth connections to each service set up separately, or does it somehow inherit existing auth from apps already on the phone? That setup friction is usually where "without leaving the conversation" tools lose people before they ever see the value.
How does Acti avoid opening the wrong document or calendar event when multiple items have similar names?
This is really handy. I actually use AI agents for work too, and sometimes I give them instructions from my phone using the built-in voice typing. But the problem is, it doesn't always catch what I really mean, and it ends up executing commands based on the wrong interpretation, which has been pretty annoying.