---
id: CVE-2026-17557
title: >-
  lm-evaluation-harness: lm-evaluation-harness: SSRF via job-spec model.url with
  bearer credential forwarding (CVE-2026-17557)
summary: >-
  A flaw was found in the `lm-evaluation-harness` component of Red Hat OpenShift
  AI. An authenticated user, able to submit an evaluation job, could exploit
  this by providing a malicious model URL in the `JobSpec`. This vulnerability
  allows f…
severity: medium
cvss: 6.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N'
cvssSource: vendor
cwe: CWE-918
vendor: Red Hat
product: Red Hat OpenShift AI (RHOAI)
affected:
  - openshift_ai_rhoai
published: '2026-10-21'
updated: '2026-10-21'
sourceUpdated: '2026-10-21T09:30:36+00:00'
source: CSAF
sourceUrl: 'https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-17557.json'
references:
  - url: >-
      https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-17557.json
  - url: 'https://access.redhat.com/security/cve/CVE-2026-17557'
  - url: 'https://bugzilla.redhat.com/show_bug.cgi?id=2507452'
  - url: 'https://www.cve.org/CVERecord?id=CVE-2026-17557'
  - url: 'https://nvd.nist.gov/vuln/detail/CVE-2026-17557'
tags:
  - csaf
  - vex
  - red-hat
  - cve.org
ingestedAt: '2026-10-21T08:45:36.244Z'
---

## Overview

A flaw was found in the `lm-evaluation-harness` component of Red Hat OpenShift AI. An authenticated user, able to submit an evaluation job, could exploit this by providing a malicious model URL in the `JobSpec`. This vulnerability allows for the disclosure of the tenant's model authentication secret to an attacker-controlled external listener. Additionally, it enables Server-Side Request Forgery (SSRF) attacks against internal network endpoints reachable from the evaluation job pod.

## Vendor advisories

- **Red Hat VEX** · Moderate · affected: Red Hat OpenShift AI (RHOAI) · no fix planned: Red Hat OpenShift AI (RHOAI) · updated 2026-10-21 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-17557.json)

**lm-evaluation-harness: lm-evaluation-harness: SSRF via job-spec model.url with bearer credential forwarding** — rated Moderate by Red Hat. Released 2026-10-21.

Affected:

- Red Hat OpenShift AI (RHOAI)

No fix planned:

- Red Hat OpenShift AI (RHOAI)

## Remediation

Fix deferred