Workspaces

A workspace brings related agents, skills, memories, and conversations into one shared place. It does two things at once: it scopes a set of related work into one place, and it governs who may use, maintain, and administer each part.

What a workspace gives you

Manage multiple agents at scale

Running one agent means following one conversation. Running a group means keeping the agents useful together: shared methods stay consistent, questions reach someone who can answer them, and an unattended run does not disappear from view just because nobody has its thread open.

A workspace gives that group a maintained home. Owners and Editors improve the agents and knowledge once, while Members use the shared setup. You can manage the operation across several agents without treating every conversation as an isolated setup job.

For example, Scout can triage support requests while Scribe prepares a report and Press waits for approval to publish. An Owner checks the fleet in Command Center; the person responsible for a pending request handles it in Inbox. When a reporting convention changes, an Editor updates the shared skill rather than rewriting the same instructions on every agent. Each run still has its own history, and each decision keeps its own permission boundary.

Open a workspace and get started

Use the switcher at the top of the sidebar to choose a workspace. The sidebar will update to show that workspace’s pages, and you’ll see its name at the top so you always know where you’re working.

To return to your own work, select Personal. You’ll find account and workspace settings in separate sections of the account menu.

When to reach for a workspace

Reach for a workspace when:

• Several people should maintain the same agents or skills.
• Many people should run a capability while a smaller group maintains it.
• Using, maintaining, and administering should carry different permissions.
• Context should be reviewed before people or agents rely on it.
• A set of agents should be scoped on its own, by client or by domain.

Scope

A workspace puts a wall around the agents, skills, and memories that belong together.

That boundary is useful even when you are the only person in it. Keep a workspace per client company, so the agents, methods, and reviewed facts for Acme never mix with Globex's. Keep a workspace per area of expertise, one for support, one for growth, each with its own agents and context.

What belongs to one workspace stays out of the others, and out of your personal workspace, until you decide otherwise.

Add people, and those scoped agents become a shared system. Picture a support group running three:

With agent delegation, Scout hands a resolved case to Scribe, which drafts documentation using the workspace's skills and memories, and Press publishes the approved result. The workspace holds the agents, methods, and reviewed context that make that handoff the group's, not one person's.

Governance

A workspace's whole safety model is one split: everyone can use what the workspace has, but only some can change it. Three mechanisms enforce that split.

• Roles decide who can change what. Every Member runs the workspace's agents and reads its memories; Editors maintain the shared agents, skills, and context; Owners administer the workspace itself. The smallest role that fits someone's job is the right one. See Workspace roles.

• Ownership decides who maintains a given resource. A workspace-owned agent or skill belongs to the workspace, so Owners and Editors edit it; a personal one shared in stays with its owner, who keeps the source while the workspace gets run-only access. See Workspace agents and Workspace skills.

• Review decides what becomes shared truth. Choose whether an agent saves to its own memories or the workspace library. Use suggestions and deliberate promotion when shared guidance needs human review. See Workspace memories.

Workspace membership is separate from Shared Billing.

Explore the workspace

FAQs