# CVE-2025-12805

## Summary

- **CVE ID:** CVE-2025-12805
- **Severity:** HIGH
- **CVSS Score:** 8.1 (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N)
- **CWE:** CWE-653
- **Published:** Mar 26, 2026
- **Last Modified:** Mar 31, 2026

## Description

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.

## Affected Products

- Red Hat — Red Hat OpenShift AI 2.25 (sha256:c0d95dfbae20e87113ffb81026d379bb63ad300447df98b27d1bf9a83b084744)
- Red Hat — Red Hat OpenShift AI 2.25 (sha256:1d258fe98c2477e4256a9b936f412f2501fb7ca9e3b810347f9712e0d5ce5c92)

## References

- [CNA](https://access.redhat.com/errata/RHSA-2026:2106)
- [CNA](https://access.redhat.com/errata/RHSA-2026:2695)
- [CNA](https://access.redhat.com/security/cve/CVE-2025-12805)
- [CNA](https://bugzilla.redhat.com/show_bug.cgi?id=2413101)

## Exploitation Prediction (EPSS)

- **EPSS Score:** 0.38%
- **EPSS Percentile:** 31.6

---
_Exported from OnDuty AI Vulnerability Intelligence on 2026-09-10._