Detections
Analytics

Chainguard: vllm-openai-cuda-13.0: security update to 0.31.0-r0

low Tenable Cloud Security Plugin ID 476111

Description

There are packages installed that are affected by a vulnerability referenced in the following CVE:

 - vLLM is an inference and serving engine for large language models. Prior to 0.30.0, Harmony tool
 continuations submitted through "POST /v1/responses" requests rebuild the next-turn engine input without
 preserving the cache_salt value, placing the continuation prefix in the global unsalted cache namespace
 even when the caller enabled salting. On deployments with prefix caching enabled, which is the default, an
 authenticated tenant who can reconstruct a victim's low-entropy post-tool history can submit the same
 continuation and use the cached_tokens_per_turn count to determine whether the prefix was previously
 processed, defeating the intended tenant isolation of salted prefix caching. This issue is fixed in
 version 0.30.0. (CVE-2026-105752)

Solution

Update the vllm-openai-cuda-13.0 library and its related packages to version 0.31.0-r0 or later.

Plugin Details

Severity: Low

ID: 476111

Version: Revision 1.1

Type: Local

Published: 10/17/2026

Updated: 10/17/2026

Risk Information

VPR

Risk Factor: Low

Score: 1.2

Percentile: 0.01

CVSS v2

Risk Factor: Low

Base Score: 2.1

Temporal Score: 1.6

Vector: CVSS2#AV:N/AC:H/Au:S/C:N/I:P/A:N

CVSS Score Source: CVE-2026-105752

CVSS v3

Risk Factor: Low

Base Score: 3.1

Temporal Score: 2.7

Vector: CVSS:3.0/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:N

Temporal Vector: CVSS:3.0/E:U/RL:O/RC:C

Vulnerability Information

Exploit Ease: No known exploits are available

Vulnerability Publication Date: 10/5/2026

Reference Information

CVE: CVE-2026-105752