From prodsec-skills
Secures model registry admin interfaces with authentication, RBAC, MFA, session management, and audit logging.
How this skill is triggered — by the user, by Claude, or both
Slash command
/prodsec-skills:admin-interface-securityThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
If the model registry has an administrative interface, only identified, authenticated, and authorized users MUST be able to connect to it.
If the model registry has an administrative interface, only identified, authenticated, and authorized users MUST be able to connect to it.
| Control | Description |
|---|---|
| Identification | Every admin user must have a unique identity (no shared accounts) |
| Authentication | Standard authentication mechanism (OIDC/OAuth 2.1 recommended) |
| Authorization | RBAC or ABAC to control which admin operations each user can perform |
| Multi-factor authentication | MFA recommended for administrative access |
| Session management | Short-lived sessions with automatic timeout |
| Operation | Risk |
|---|---|
| Model upload/publish | Introducing malicious or backdoored models |
| Model deletion | Removing approved models, causing service disruption |
| Access control changes | Granting unauthorized users access to models |
| Configuration changes | Weakening security settings |
| Registry metadata modification | Altering model provenance information |
npx claudepluginhub redhatproductsecurity/prodsec-skills --plugin prodsec-skillsEnforces authenticated and authorized access to model registry storage with RBAC, encryption, access logging, and network isolation controls.
Guides selection and implementation of authorization models including RBAC, ABAC, ACL, ReBAC, and policy-as-code for permission systems and access control design.
Configures Mistral AI enterprise RBAC via workspaces for team-scoped API keys, model restrictions, rate limits, and budgets. Includes bash API setup and TypeScript app enforcement.