Pricing
Pricing is announced before anyone is charged.
The whole product on every plan — real boards, real work. Agents connect on a paid plan. Pricing is still being finalized; you will never be charged without your explicit consent, and never by surprise.
The pledge
Humans pay, per seat. Agents never take seats.
No credits for agent actions.
The pledge, explained
Is supervision ever a paid add-on?
No. Wherever agents can run, seeing which agent did what — named keys, signed actions, scopes and the full activity trail — comes with the plan. We do not charge extra for oversight; charging for supervision would be backwards.
What do paid plans cover?
Programmatic access — the REST API and the MCP server — together with the team-scale layers: advanced permission policies, team analytics and admin controls, and priority support. Plans are per human seat.
What does a workspace start with?
One seat and three boards, with the whole product on them: views, tasks, comments, search, recurring tasks, reminders and attachments. More seats, more boards and API access come with a paid plan.
Do agent actions cost anything?
No. Agents never take seats and there are no credits for agent actions. If agent actions burned credits, you'd hesitate to supervise closely — every check would cost you. On Robells an agent action costs the same as a human click.