The apiary for teams

On your own, the apiary remembers your work across sessions and tools. On a team, it does something more valuable: what one person learns can reach everyone else.

#What changes when you add a team?

A teammate solves a tricky migration on Monday, and the reusable lesson is available to the whole team's assistants by their next session, without anyone passing around a file or a Slack message. That is the core benefit of team mode: one brain instead of several disconnected ones.

#How does memory stay separated across projects and teams?

Teams need memory that is shared and separated at the same time, so the apiary organizes it into three nested levels.

Org is your company. It is the outer boundary, two different companies never see each other's anything.

Workspace is a team within the company. Two teams keep separate memory, enforced at the storage layer, not just inside the app.

Project is the specific repository or folder you are working in. Within a team, memory is scoped to the project you are actually in, so a note from one repo does not surface while you work in a different one.

You do not manage the project level by hand. The apiary figures out which project a session belongs to from the folder you are working in, even across several open projects at once in different tools.

#What is shared and what is private?

Inside a workspace, what an assistant can see depends on a simple policy:

• Private: an assistant sees only its own memories. Good for a personal or CI assistant that should not mix into the shared pool.
• Shared: an assistant sees the team's shared memories plus its own. This is the one-brain setting.
• Group: an assistant shares with a named group of teammates, plus its own.

The default leans private and safe. When in doubt, the apiary shows less, not more. You widen sharing on purpose, never by accident.

#How does a teammate's discovery reach me?

Shared skills are the most visible team benefit. When a skill is published to the team, every teammate's assistants pick it up at the start of their next session. Skills carry who wrote them, so credit and history are clear, and two people can name a skill the same thing without one clobbering the other. Promoting something from "just mine" to "the whole team's" is always a deliberate, recorded action, so nothing private gets shared by surprise.

#How do I switch between projects, teams, and companies?

Because the apiary scopes to the folder you are in, moving between repositories needs no manual switch at all, just open the other project in your assistant and it follows. Moving between the teams and companies you belong to stays a deliberate choice:

honeycomb org list
honeycomb workspace list
honeycomb project list
honeycomb org switch acme
honeycomb workspace use backend

The dashboard offers the same switches in a menu, showing only the orgs, workspaces, and projects you actually have access to.

#Whose store is it?

A team's memory lives in your own Deeplake store, with each team and project separated at the storage layer, and you can host that storage in your own cloud account if you want. Sensitive credentials, like API keys, are never stored alongside memory and are never shown to an assistant. Memory is versioned and inspectable, sharing is opt-in by design, and nothing leaves your store except sign-in traffic and, only if you allow it, anonymous usage counts.

#Common questions

#How does a teammate's discovery reach me?

When something reusable is published to the team as a skill, every teammate's assistants pick it up automatically at the start of their next session. Nobody copies files around.

#Will my private notes get shared by accident?

No. The default leans private, an assistant sees only its own memories unless sharing is widened on purpose. Promoting something from personal to team-wide is always a deliberate, recorded action.

#How does memory stay separated across projects and teams?

The apiary organizes memory by org (company), workspace (team), and project (the repository or folder you are in), enforced at the storage layer. It scopes to the project you are actually working in automatically, so switching repos needs no manual action.