Super Agent privacy, security, and permissions

Super Agents are human-like, agentic teammates in your Workspace. Super Agents are treated as ClickUp users. This allows us to leverage permissions to control Super Agent access to your Workspace data.

This article explains how security, permissions, and access work for Super Agents. Learn about our privacy and security policies for our AI features as a whole.

Feature availability and limits vary by plan and user role. Learn more

There are privacy, security, and permissions that only apply to Super Agents. Learn more

To use Super Agents, you'll need to use ClickUp 4.0 and activate the AI ClickApp for your Workspace.

If you want to interact with Super Agents via Chat, you'll need to activate the Chat ClickApp.

Super Agents are managed by people

When you create a Super Agent, you configure the ways they can interact with your Workspace and teammates. 

For example, if a Super Agent does not have the Docs toolset, it can't create Docs and pages or edit existing Docs. 

Anyone with Can manage permissions can edit a Super Agent's configuration at any time. 

Super Agents collaborate alongside people

There is always a person in the Super Agent workflow.

If a Super Agent's trigger is being @mentioned, it can't be triggered in any other way. Unless the Super Agent is triggered, it is not active in the Workspace. If you want a Super Agent to check with a person before taking action, you can tell it to do so by configuring the Super Agent's instructions or preferences.

Super Agent's access to Workspace data

So that Super Agents can be truly ambient teammates, they are required to have access to all public locations and items in your Workspace.

Implicit access to Workspace data

Once created, Super Agents automatically receive public data from the Workspace as it's added. This means that they always respond with up-to-date information. 

Explicit access to Workspace data

When creating a Super Agent, you can manually give it access to specific Workspace locations using the knowledge section of its profile. 

Super Agents' use of knowledge

When a Super Agent is mentioned in a private location or a location that was not selected in its knowledge, the Super Agent has temporary access to the data in that location. This means that the Super Agent can see information and perform actions within that location only for the duration of that interaction.

If Intelligence is activated in a Super Agent's profile, it can store information from private locations in its memory.

When you select a location in the Super Agent's knowledge, or explicitly give it access as part of its instructions, it has persistent access to the data in that location.

Workspace user access to Super Agents is controlled

We've implemented layers of default and editable permissions for users on the Workspace and individual levels. You can make your Super Agents private or share them with specific people or Teams. 

If private or external data is shared with the Super Agent, it can use that data when responding to anyone who has permission to trigger the Super Agent, even if they don't have access to the data directly. Responses, or any other content Super Agents generate, are visible to anyone with access to that response or content.

Default Workspace-level permissions

These are the default Workspace-level permissions:

Guests and limited members: 

  • These roles cannot see Super Agents' profiles or create or trigger a Super Agent. 
  • In locations where they have access, these roles can see Super Agent comments, generated content, and avatars, but they can't respond to them or open their profile. 

Members, admins, and owners: 

  • By default, these roles can create, trigger, manage, and interact with all public Super Agents. 
  • Any private info in the Super Agent's profile that they do not have access to will be hidden from them.

Editable permissions

These are the editable permissions:

Private Super Agents

Super Agents can be made private. Private Super Agents are only visible to the person who created them, but all Super Agents can be shared with individual users.

Transparent privacy handling

When configuring a Super Agent, you will see a warning if there are any privacy risks to ensure you're aware.

Private info can be exposed to anyone who has permission to trigger a Super Agent. Here's an example of how this works:

Avery adds a private List to their Super Agent's knowledge. The Super Agent now has access to that List. Riley doesn't have access to the private List. Riley asks the Super Agent a question in a private Chat Channel, and it uses info from that private List to answer. Alex has no access to the private Super Agent or the private List. Alex has access to the private Chat Channel and can see the Super Agent's response to Riley.

Super Agent audit log

There is a Super Agent audit log for troubleshooting and auditing purposes. There is also a Workspace audit log that records Workspace users' Super Agent activity.  

Super Agent memory

Check out our article explaining the permissions and security included in Super Agent memory.