One app for every agent. Change with a tap.

Each AI company builds an app for its own agent, with its own habits. Ripul puts them behind one set of screens, so what you learn with one carries over to the next.

One list of conversations

Every agent's work in one place, named from what you asked for, so the list reads like a to-do list.

One way to say yes

The same approval question, whichever agent is asking.

One kind of notification

Finished, needs you, or stuck. The same three, for all of them.

The best agent changes from month to month, and sometimes from job to job. Changing agent or model is a tap inside a running conversation, and the conversation carries on.

Pin the two or three you use most and they become one-tap buttons for starting a new conversation. A good habit is to begin with a quick, light model and move to a heavier one only when the job turns out to need it.

One agent per job, all of them working while you read one of them. Start one on the problem you are chasing, another on the tidy-up you have been putting off, and a third on a different project altogether.

The list shows which are working, which are waiting for you and which have finished. A conversation that needs your answer brings itself to the top.

Every long conversation has a moment where it went wrong. You do not have to argue with it or start again.

Fork

Branch from any point into a new conversation. The original carries on, so you can try two approaches and compare.

Restart

Run again from a chosen message with a better instruction, without the wrong turn.

Resume

Pick a conversation up days later with everything it knew still in place.