Who can do what.
Real control for growing teams. People and AI agents, governed the same way.
Members & roles
Give your agency view-only access to one folder. Let a contributor edit but not publish. You decide who sees and does what, in plain choices.
Access is a ladder of plain choices, from viewing content only to managing the brand, and each rung says in one line what it lets a person do. Group people into teams and give a whole team access in one move. Set your brand's languages once and every module follows. It feels more like sharing a doc than configuring a server.
- Access as a ladder of plain choices, each with what it lets you do
- Teams: give a group access in one move
- Access set per folder and per item when you need it
- Every change on the record
These choices gate every module, and they are the same rules your AI agents live under.
AI agents
AI agents are members here. Add one, choose what it can do, and hand it a key. From then on you can see everything it does, and on whose behalf.
Adding an AI agent feels like inviting a colleague: name it, tick the boxes for what it can do, and you're done. No terminal, no jargon. Your own Claude can connect to the brand and work inside it, with exactly the access you choose, and every action it takes lands on the record of the item it touched.
- Add an agent with “what this agent can do” checkboxes
- Agents show up as members
- Acts on a person's behalf, never with more access
- Every action on the record: who, for whom, when
It runs on the same members and access your team already uses, so the control and the record are identical. That is also what makes it ready for the EU AI Act: the provenance is part of the platform, not an add-on.
Start with everything.
Every module, on every plan. A 30-day free trial, no credit card, no sales call.