Standard Rival User
Goal: get work done independently. The self-serve, non-enterprise baseline. Functions like an independent user with full personal freedom over tools, models, and billing.
Roles overview
Permissions are how Rival Enterprise provides your organization with the speed of automated AI without sacrificing control, auditability, or compliance. Every user in Rival holds a specific role that governs what actions they can perform, what data and tools they can see, which models they can query, and which system operations require administrative approval before completion.
The three core roles
Section titled “The three core roles”Rival structures platform access across three primary roles designed for distinct operational goals:
Enterprise Admin
Goal: manage and govern AI across the organization. IT, security, and operations leadership. Defines global guardrails, manages user lifecycles, configures connectors, and maintains complete system visibility.
Enterprise Team Member
Goal: execute daily work within company guardrails. Enterprise employees. Actively builds and runs automations using approved tools, vetted connectors, and required approval checkpoints.
For the granular, action-by-action breakdown, see the Capability matrix.
Core access hierarchy
Section titled “Core access hierarchy”The platform’s permission structure rests on a clear division of authority:
- Standard Rival Users can execute any operation for themselves in an unmanaged workspace, taking on full responsibility for their own data, credentials, and billing.
- Enterprise Admins establish global security rules, approve marketplace asset catalogs, manage team budgets, and enforce model guardrails.
- Enterprise Team Members operate freely and efficiently inside those pre-established administrative boundaries without needing to hand-code security checks or manage credentials.
Detailed role breakdowns
Section titled “Detailed role breakdowns”Standard Rival User
Section titled “Standard Rival User”Designed for standalone builders and individuals operating outside an enterprise workspace boundary:
- Asset creation — full authority to build, edit, and publish custom functions, agents, and workflows for personal use.
- Model selection — unrestricted access to any supported foundation model available on the platform.
- Resource management — direct management of individual run balances, personal API keys, and private system connections.
Enterprise Admin
Section titled “Enterprise Admin”Designed for IT, security, and department leadership requiring full administrative command over organizational AI adoption:
- Governance and policies — configure global model guardrails, restrict PII transmission, and define approved marketplace catalogs. See Guardrails and allowed models.
- Access and provisioning — manage SSO and SCIM integrations, invite users, organize teams, assign roles, and revoke access instantly. See User lifecycle.
- Financial oversight — set departmental cost controls, track usage telemetry across teams, and allocate run balances centrally. See the Cost Optimization Center.
- Approval gates — review and approve pending marketplace submissions, connector access requests, and high-consequence workflow actions. See Approval workflows.
Enterprise Team Member
Section titled “Enterprise Team Member”Designed for employees executing business tasks and building team-level automations inside governed environments:
- Approved capabilities — access and execute all admin-vetted tools, agents, and workflows published to the internal marketplace catalog.
- Guided creation — construct new custom tools and agents using visual builders or RivalBot, which automatically submit to the admin queue upon publication.
- Scoped integrations — interact with central enterprise connectors explicitly granted to their team or role. See Connector governance and security.
A quick example: governed automation lifecycle
Section titled “A quick example: governed automation lifecycle”Consider how roles interact when deploying a new customer support automation:
- Rule definition — an Enterprise Admin configures global guardrails blocking the transmission of customer PII and approves specific support connectors.
- Safe execution — an Enterprise Team Member builds an agent to draft ticket responses, operating smoothly within the approved connectors and rules.
- Escalation gate — when the team member’s agent attempts a high-impact operation, such as issuing a customer refund, the system pauses execution automatically and routes an approval request to the Enterprise Admin.
Role capability comparison
Section titled “Role capability comparison”| Operational area | Standard Rival User | Enterprise Admin | Enterprise Team Member |
|---|---|---|---|
| User and team management | — | Full control | View team members |
| Tool and agent creation | Unrestricted | Unrestricted | Guided / governed |
| Publishing to org catalog | Direct (personal) | Direct (org-wide) | Requires admin approval |
| Connector access | Personal accounts | Central setup and assignment | Assigned / approved only |
| System audit logs | Personal only | Global enterprise visibility | Team-scoped activity |
Next steps
Section titled “Next steps”- Review the granular capability breakdown → Capability matrix
- Learn how interface views adapt per role → Navigation by role
- Manage user invitations and team assignments → Users and invitations