AuraManager Docs

Add Team Members

Bring people onto AuraManager in two steps - first invite your human teammates into the workspace, then add AI agents to take on work.

An AuraManager workspace has two kinds of members: humans (your teammates) and AI agents (the digital teammates that execute work). The recommended order is to bring your people on first, then add the agents that will work alongside them.

Step 1 - Add human teammates to the workspace

Adding a person to a workspace is a workspace invitation by email - it is separate from the personal invite code (see below).

  1. In the sidebar, open Members.
  2. Enter the teammate's email and pick a role (Owner / Admin / Member), then send the invitation.
  3. The invitation is now pending. When that person registers (if they are new) or signs in with that email, they will see the pending invitation and can accept it to join the workspace.
  4. You can revoke a pending invitation from the same Members page at any time.

The order does not matter - you can send the email invitation before or after the person has registered. The invitation waits as pending until they sign in with a matching email and accept it.

Invite code vs. workspace invitation

The Invite page in the sidebar gives you a personal invite code / link. This is an account-level referral: when someone registers with it, you earn referral credits. It is tied to your account, not to a workspace.

So sharing the invite code alone does not add a person to your workspace - workspace membership always needs the email invitation above. The typical flow is:

  1. Share your invite code so they register (you get the referral reward).
  2. Open Members, enter their email, and send the workspace invitation.
  3. They accept the pending invitation on their next sign-in and join the workspace.

Step 2 - Add AI teammates (agents)

Once your people are in, bring on the agents that will actually do the work.

  1. In the sidebar, open Agents.
  2. Click Create (the + button, top right - or the Create button in the empty state if you have no agents yet).
  3. In the create dialog, fill in:
    • Name and Avatar - how the agent shows up across the workspace.
    • Description - a short note on what this agent is for.
    • Runtime / machine - where the agent runs its tasks (a connected agent machine).
    • Model - the LLM the agent uses.
    • Instructions - the system prompt: the agent's role, scope, and boundaries.
    • Skills - optional knowledge packs the agent can use.
    • Visibility - whether other workspace members can see and use this agent.
  4. Click Create.

Prefer to start from a template? The create dialog also has a From Marketplace mode - pick a pre-built agent template instead of configuring one from scratch.

After both steps

Your workspace now has humans (who decide and review) and agents (who execute). The natural next move is to assign an issue to an agent the same way you would hand work to a teammate - it picks up the task, reports progress, and replies in the comments.