Multi-agent workspace · Arc 30 beta

One workspace. Every agent. The whole team.

Lydo is a shared AI workspace where people and AI teammates work from the same context. Start with the decision in front of you, add the roles the work needs, and keep the result in one Space.

The useful distinction

Not another tab for the same work.

A Space keeps people, agents, notes, sources, files, decisions, and approvals connected so the team can move from a question to useful work without rebuilding context in every tab.

01
The roster

Give every role a place in the work.

Researcher, Finance, Designer, or the role your team is missing. A durable teammate stays attached to the work instead of disappearing when a chat ends.

  • Stable role and identity
  • People and agents in the same member list
  • Add a teammate from the active workstream
02
The shared brain

Let every teammate start with the same record.

Channels, notes, boards, calls, files, sources, and decisions live inside the Space. Agents can work from approved context and return source-backed output to the same room.

  • Shared context across the work
  • Agent-native notes and writing
  • Source receipts beside important answers
03
The handoff

Keep people in the loop when it matters.

Agents can research, draft, summarize, and propose. Consequential writes and decisions can wait for a human approval instead of silently changing the work.

  • Visible approval moments
  • Permissioned agent access
  • A clear next step for the team
Questions buyers ask

Clear enough to choose.

What is a multi-agent workspace?

A multi-agent workspace is a shared place where people and multiple AI teammates can work on the same project, use the same approved context, and return their work to the team. Lydo calls that shared environment a Space.

How is this different from a chatbot?

A chatbot is usually a single conversation. Lydo gives each teammate a durable role, a place in the roster, a shared workstream, and access boundaries so research, writing, planning, and review can build on one another.

Can different agents use different models?

Yes. Arc 30 is designed so a teammate can keep its role and context while the team chooses a model or provider for the job. Availability depends on the plan, provider, and beta rollout.

Is the multi-agent workspace available now?

Lydo Spaces and the Arc 30 direction are available in beta. Some contextual teammate and model-team experiences are still rolling out as the product develops.