vLLM: Dependency Confusion Vulnerability in vLLM Dockerfile

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.

Problem type

CWE-427: Uncontrolled Search Path Element

Affected products

vllm-project

vllm

< 0.22.1 - AFFECTED

References

https://github.com/vllm-project/vllm/security/advisories/GHSA-jrf6-vqxq-pjv2

https://github.com/vllm-project/vllm/security/advisories/GHSA-jrf6-vqxq-pjv2

x_refsource_CONFIRM

JSON source

https://cveawg.mitre.org/api/cve/CVE-2026-54232

Click to expand

{
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "cveMetadata": {
    "cveId": "CVE-2026-54232",
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "dateUpdated": "2026-06-22T22:16:43.101Z",
    "dateReserved": "2026-06-12T16:25:43.084Z",
    "datePublished": "2026-06-22T22:16:43.101Z",
    "state": "PUBLISHED"
  },
  "containers": {
    "cna": {
      "providerMetadata": {
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M",
        "dateUpdated": "2026-06-22T22:16:43.101Z"
      },
      "title": "vLLM:  Dependency Confusion Vulnerability in vLLM Dockerfile",
      "descriptions": [
        {
          "lang": "en",
          "value": "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."
        }
      ],
      "affected": [
        {
          "vendor": "vllm-project",
          "product": "vllm",
          "versions": [
            {
              "version": "< 0.22.1",
              "status": "affected"
            }
          ]
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "lang": "en",
              "description": "CWE-427: Uncontrolled Search Path Element",
              "cweId": "CWE-427",
              "type": "CWE"
            }
          ]
        }
      ],
      "references": [
        {
          "url": "https://github.com/vllm-project/vllm/security/advisories/GHSA-jrf6-vqxq-pjv2",
          "name": "https://github.com/vllm-project/vllm/security/advisories/GHSA-jrf6-vqxq-pjv2",
          "tags": [
            "x_refsource_CONFIRM"
          ]
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "version": "3.1",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H",
            "attackVector": "NETWORK",
            "attackComplexity": "LOW",
            "privilegesRequired": "NONE",
            "userInteraction": "REQUIRED",
            "scope": "UNCHANGED",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "availabilityImpact": "HIGH",
            "baseScore": 8.8,
            "baseSeverity": "HIGH"
          }
        }
      ]
    }
  }
}