CVE-2025-12805
A flaw was found in Red Hat OpenShift AI (RHOAI) llama-stack-operator. This vulnerability allows unauthorized access to Llama Stack services deployed in other namespaces via direct network requests, because no NetworkPolicy restricts access to the llama-stack service endpoint. As a result, a user in one namespace can access another user’s Llama Stack instance and potentially view or manipulate sensitive data.
Scoring
- Severity
- HIGH
- CVSS base score
- 8.1
- CVSS vector
- CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N
- EPSS probability
- 0.38%
- CWE
- CWE-653
- Published
- 2026-03-26
- Last modified
- 2026-03-31
Affected products
- Red Hat Red Hat OpenShift AI 2.25
- Red Hat Red Hat OpenShift AI 2.25
Weakness type
Related vulnerabilities
- CVE-2026-15366 — A control logic defect in a specific built-in webpage of Kids Mode allows users to view local...
- CVE-2026-71325 — Traefik: `allowCrossNamespace=false` bypass via `@kubernetescrd` TraefikService backendRef
- CVE-2026-62246 — Kamaji: TenantControlPlane namespace/name collision binds two tenants to the same SQL datastore schema + DB user, breaking per-tenant isolation
- CVE-2026-65635 — Boruta dynamic client registration allows creation of over-privileged OAuth clients
- CVE-2026-53421 — Apache Syncope: Remote Code Execution via Scripted Connector
- CVE-2026-53405 — Apache Syncope: Remote Code Execution via Flowable BPMN Groovy ScriptTask
- CVE-2026-63071 — Apache Syncope: RCE via Groovy Sandbox bypass
- CVE-2026-15738 — Cross-namespace traffic interception via incorrect route precedence ordering in AWS Load Balancer Controller