CVE-2025-12805 PUBLISHED

Llama-stack-k8s-operator: llama stack service exposed across namespaces due to missing networkpolicy

Assigner: redhat
Reserved: 06.11.2025 Published: 26.03.2026 Updated: 26.03.2026

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.

Metrics

CVSS Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N
CVSS Score: 8.1

Product Status

Vendor Red Hat
Product Red Hat OpenShift AI 2.25
Versions Default: affected
  • unaffected from sha256:c0d95dfbae20e87113ffb81026d379bb63ad300447df98b27d1bf9a83b084744 to * (excl.)
Vendor Red Hat
Product Red Hat OpenShift AI 2.25
Versions Default: affected
  • unaffected from sha256:1d258fe98c2477e4256a9b936f412f2501fb7ca9e3b810347f9712e0d5ce5c92 to * (excl.)
Vendor Red Hat
Product Red Hat OpenShift AI (RHOAI)
Versions Default: unaffected
Vendor Red Hat
Product Red Hat OpenShift AI (RHOAI)
Versions Default: unaffected

Workarounds

Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base or stability.

References

Problem Types

  • Improper Isolation or Compartmentalization CWE