Every team we knew was answering the same dozen questions by hand, and every tool offered to fix it by guessing. So we built one that answers from your own content, admits it when your content does not cover something, and hands that case to a person — in the system your team already opens every morning.
Support chat went through a strange few years. The tools got fluent long before they got honest, and fluency is exactly the wrong thing to be good at first. An agent that invents a refund window, a delivery time or a policy does not save your team an hour — it costs them a day, and it costs a customer their trust in whatever they read next.
We think the useful version is narrower and much more boring. It reads what you have already written — your pages, your files, your policies, the tickets sitting in your own database — and it answers from that or not at all. When your content does not cover the question, it says so in the words you chose, collects what your team will need, and raises a ticket.
The second thing we kept running into was where that ticket went. Nearly everything wanted to become the place your team works: another inbox, another board, another login, another thing to check. Nobody needs that. Your tickets should land in the system you already run, with your fields and your workflow, and the agent should be the quiet part at the front.
An honest “I don’t know, let me get someone” is worth more than a confident paragraph that is wrong.
Not positioning. These are the decisions the product is actually built around, and the ones we would argue with a customer about.
The agent answers from the content you gave it. Where that runs out it does not improvise — it falls back to the sentence you wrote for exactly that case and offers to fetch a person. You can read every answer it gave and see which chat it came from.
Database sources are read over a read-only connection — we cannot write to your data even by accident. Emails, phone numbers and card-shaped numbers are masked before text reaches the model, and files a visitor sends sit in a private bucket behind short-lived links.
Tickets are posted to an endpoint in your own project, as JSON, with the extra fields you define. Your database, your workflow, your reports. One fewer tab for your team, and nothing of yours locked inside something of ours.
Every AI call is recorded against the chat it ran for, with its tokens and its cost, so a month is a number you can read rather than one you find out about. Plans are priced per website, not per person, so nobody on your side has to ration seats.
The agent is only as good as what it is allowed to read and where it is allowed to put the result. So most of the work has gone into those two ends rather than into the conversation in the middle.
What it doesAnything the agent cannot answer becomes a ticket in your system — not in ours.
We are a small software team in India. We build the product, we run the servers it sits on, and when you write to us it is one of us who writes back — not a queue, and not an agent answering about the agent.
Add a website, point it at a page you already have, and ask it the question your team answers most often. An afternoon will tell you more than this page can.