One agent you talk to. A team behind it.
Roy is the executive agent. You talk to Roy. Behind him is a named team of specialists, each with their own domain and their own memory, and Roy decides which of them a question belongs to.
Why a team beats one chat window
A single context holding everything you do is a single context that is wrong about most of it.
A specialist keeps its own memory
The agent that watches your pipeline remembers every deal and every touch. It is not competing for room with your house plan or last quarter's board paper.
You ask once
Ask a question that crosses three domains and Roy fans it out, collects the answers, and comes back with one. You do not pick the right tab first.
Separation is enforced, not promised
Specialists hold separate stores. The health agent cannot read your deals. That is a property of how the team is built, not a setting.
How work gets routed
- You ask Roy. One place, in your words. No routing decision to make and no syntax to learn.
- Roy picks the specialists. A question about your market reaches Marlowe and the research desk. A post to write reaches Ellis. A change to Roy itself reaches the builder.
- They answer from their own memory. Each specialist brings what it has been watching, with the evidence attached.
- Roy returns one answer. Reconciled, cited, and with a proposed next step you can approve or throw out.
The roster
Each one has a name so you can read its thread. All software, all gated, all handing their work to the same queue.
Give it a week. It'll know your job.
Private beta is open to a small group of operators. Tell us what you run and we'll get you in.
Get early access