Permissions matrix
What a user can do with dashboard comments depends on their
access level to the dashboard’s workbook.
Admin roles are billed at the developer rate.
Restricting data downloads
By default, all roles can download query results as CSV from workbooks, Analytics Chat, and published dashboards. Two controls restrict this:- Account-wide — the Allow data downloads switch on the Admin → Settings page. Turning it off hides export controls everywhere, for all users — including admins and embeds.
- Per user — the Download data action in custom roles. Built-in roles grant it by default; to remove download access for a group of users, assign them a custom role without it. Admins and anonymous embed viewers bypass it.
Agent Permissions
Agents are connected to Semantic Model Deployments and inherit the permission level of the user they are operating under. Each agent can be configured to use the Restrict views feature which allows to select the views that are visible to the agent. This feature should not be used as a security measure. Configure access policies instead.Typical Usage Scenarios
- Viewers: Business users who consume published dashboards, use Analytics Chat, and query data through external tools like Tableau or Power BI
- Explorers: Typically data consumers and analysts
- Developers: Usually data stewards and data engineers
- Admins: Typically assigned to data engineers managing the entire Cube instance (billed at the developer rate with additional privileges)
Built-in role IDs
The REST API and the CLI refer to roles by numeric ID. Built-in roles have fixed negative IDs, the same in every Cube account:
Admin cannot be assigned by ID:
PUT /api/v1/roles/{roleId}/users/{userId} answers 404 for it. Grant it to a user with the isAdmin flag on PUT /api/v1/users/{id}, or through a group.
Users created before the current role set may hold the legacy Developer role, ID
-1. It is the predecessor of -3, and the console replaces it whenever a user’s role is changed there. When replacing Developer over the API, handle -1 as well as -3: both IDs work with the calls below.Replacing a built-in role
Permissions stack, so giving a user a custom role does not take away what a built-in role grants. To move a user from Explorer to a custom role, assign the custom role first, then remove Explorer:PUT and DELETE on /api/v1/roles/{roleId}/users/{userId}. Both are idempotent, so a script can run them for every user and be re-run after a failure.
To find who still holds a role, call GET /api/v1/roles/-4/users. It returns users newest first in items; pass first for the page size and, while pageInfo.hasNextPage is true, pass pageInfo.endCursor as after for the next page. Each user carries roles, every role they hold, so “holds Explorer but not the custom role yet” is a filter over those pages:
GET /api/v1/users/{userId}/roles returns the same isAdmin and roles for one user. In both, isAdmin is true whether Admin comes from the user’s own flag or from a group. Deactivated users are included, so a swap reaches them too before anyone reactivates them.
direct: true marks a role assigned to the user, which DELETE removes. A role held through a group shows direct: false and the granting groupIds; DELETE leaves it in place. Remove it from the group in the console instead, which revokes it from every member of that group.
All of these calls are admin-only and need an account-wide API key: a key scoped to specific deployments answers 403.
Single sign-on never removes roles: a later sign-in only adds the roles your identity provider maps, so a role removed this way stays removed unless a mapping grants it again.