8.8
/ 10
HIGH
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H
Description
vLLM is an inference and serving engine for large language models (LLMs). Prior to 0.22.1, the vLLM Dockerfile is vulnerable to a dependency confusion attack through the flashinfer-jit-cache package. The package is installed from a custom index (flashinfer.ai/whl/) using --extra-index-url, but the package name was not registered on PyPI, and UV_INDEX_STRATEGY="unsafe-best-match" is set globally. An attacker who registers flashinfer-jit-cache on PyPI with version 0.6.11.post2 can execute arbitrary code as root during the Docker build and backdoor every resulting container image, enabling exfiltration of all user prompts, API credentials, and model data from production vLLM deployments This vulnerability is fixed in 0.22.1.
AI Analysis
Dependency confusion vulnerability in vLLM Dockerfile allowing arbitrary code execution as root
Basic Information
ID
CVE-2026-54232
Source
GitHub_M
Published
Jun 22, 2026 at 22:16
Affected Product
Vendor
vllm-project
Product
vllm
Version
< 0.22.1
Affected Versions
vllm-project vllm < 0.22.1
CWE Classification
AI Assessment
AI Score
8.8 / 10
AI Severity
High
Vendor
vllm-project
Product
vLLM
Version
< 0.22.1