Your app, more personal. One agent for every user.
Each user gets an agent with memory, your knowledge, and your app’s actions. Your team sets per-user caps and overall spend limits.
Early access for the teams building what’s next.
In early access. Agents and memory are live. This demo also previews actions, knowledge, spend controls and analytics.
You set the rules.
For every agent in your app.
Your provider’s prices. No markup.
One new conversation.
SAVED BETWEEN CONVERSATIONSMAYDinner for two. One vegetarian.JUNEMild dishes, every time.AUGUSTNeighbourhood Kitchen is a favourite.NEXT CONVERSATIONAlready knows your user.
What your users gain.
What your team controls.
You add a branded chatbox to your website or app. Your users stay in their account. Your team runs the agents from one console.
Try the complete experienceYour users return without starting over.
Memory carries their preferences and history into the next conversation.
Their request becomes an action.
Add the plan, prepare the basket, or fill checkout using what your app already does.
Your team sets the spending rules.
Set a cap per user and a shared monthly budget, with alerts and behavior at the limit.
The answers come from your product.
Connect your own catalogue, policies and content so the agent draws on what your business knows.
Less searching.
More getting somewhere.
Your user opens twelve menus, checks the vegetarian options, then starts again for their partner.
Finding dinner becomes the evening’s task.
Your users say what dinner needs to be.
Your app puts two suitable meals in their basket, ready to review.
Meet the workspace
behind the conversation.
Manage your agents, correct user memory, and set per-user and overall spending limits.
This is your team’s workspace. End users use the chatbox in your existing website or app.
Switch environments here, then explore the full console on the Product page.
Next month,
no introductions needed.
In your learning app, a user returns after a month away. Their agent picks up with the trip to Paris and the listening practice they wanted.
Your app doesn’t make them start over.
“I’m going to Paris in October. Ten minutes a day is realistic. Listening is the hard part.”
“Can we pick this back up?”
“Of course. Let’s practise ordering at a café for your October trip to Paris. We’ll keep it to ten minutes, with a little more listening.”
Set the limits
before the conversations start.
Set how much each person’s agent can spend, then set a monthly cap for the environment as a whole. Your team controls both the individual usage and the total.
Choose alerts and what happens at a limit: refuse new requests or switch to a cheaper model.
Explore limits & spendThe shared budget for your production environment.
Each user’s agent works within the limit your team sets.
Your model bill
stays yours.
Connect your own provider account. Your agents use your model, and your provider bills you directly at their prices.
How pricing worksThe account you already use.
A separate bill, based on active end users.
More conversations don’t turn one person into more users.
The reassuringly boring part.
A little more detail ↗01Production and staging environments.
02Published agent versions never change.
03Memory you can read and correct, per user.
04Nothing is ever hard-deleted.