CVE-2026-5241 (GCVE-0-2026-5241)

Vulnerability from cvelistv5 – Published: 2026-06-03 12:33 – Updated: 2026-06-03 15:45
VLAI
Title
Policy Bypass in LightGlue Nested Config Resolution in huggingface/transformers
Summary
A vulnerability in the LightGlue model loading path of huggingface/transformers version 5.2.0 allows an attacker-controlled model repository to execute arbitrary code during model initialization. The issue arises because the `trust_remote_code` parameter, intended to prevent remote code execution, is overridden by untrusted serialized configuration data in a nested code path. Specifically, when loading a LightGlue model using `AutoModel.from_pretrained()` with `trust_remote_code=False`, the `LightGlueConfig` reads the `trust_remote_code` value from the untrusted `config.json` file and propagates it into nested `AutoConfig.from_pretrained()` calls. This results in the execution of attacker-provided Python modules, even when the victim explicitly disables remote code execution. The vulnerability poses a high risk for environments such as API inference servers, research notebooks, CI/CD pipelines, and model evaluation workers, potentially leading to credential theft, lateral movement, or persistence/backdoor deployment.
Severity
8 (High)
CVSS:3.0/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:N
SSVC
Exploitation: poc Automatable: no Technical Impact: total
CISA Coordinator (v2.0.3)
CWE
  • CWE-829 - Inclusion of Functionality from Untrusted Control Sphere
Assigner
References
Impacted products
Vendor Product Version
huggingface huggingface/transformers Affected: unspecified , < 5.5.0 (custom)
Create a notification for this product.
Show details on NVD website
To clipboard 

{
  "containers": {
    "adp": [
      {
        "metrics": [
          {
            "other": {
              "content": {
                "id": "CVE-2026-5241",
                "options": [
                  {
                    "Exploitation": "poc"
                  },
                  {
                    "Automatable": "no"
                  },
                  {
                    "Technical Impact": "total"
                  }
                ],
                "role": "CISA Coordinator",
                "timestamp": "2026-06-03T15:45:14.993177Z",
                "version": "2.0.3"
              },
              "type": "ssvc"
            }
          }
        ],
        "providerMetadata": {
          "dateUpdated": "2026-06-03T15:45:36.684Z",
          "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
          "shortName": "CISA-ADP"
        },
        "references": [
          {
            "tags": [
              "exploit"
            ],
            "url": "https://huntr.com/bounties/ceb3ce1a-4c45-497a-b25e-cb9a7685e619"
          }
        ],
        "title": "CISA ADP Vulnrichment"
      }
    ],
    "cna": {
      "affected": [
        {
          "product": "huggingface/transformers",
          "vendor": "huggingface",
          "versions": [
            {
              "lessThan": "5.5.0",
              "status": "affected",
              "version": "unspecified",
              "versionType": "custom"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "A vulnerability in the LightGlue model loading path of huggingface/transformers version 5.2.0 allows an attacker-controlled model repository to execute arbitrary code during model initialization. The issue arises because the `trust_remote_code` parameter, intended to prevent remote code execution, is overridden by untrusted serialized configuration data in a nested code path. Specifically, when loading a LightGlue model using `AutoModel.from_pretrained()` with `trust_remote_code=False`, the `LightGlueConfig` reads the `trust_remote_code` value from the untrusted `config.json` file and propagates it into nested `AutoConfig.from_pretrained()` calls. This results in the execution of attacker-provided Python modules, even when the victim explicitly disables remote code execution. The vulnerability poses a high risk for environments such as API inference servers, research notebooks, CI/CD pipelines, and model evaluation workers, potentially leading to credential theft, lateral movement, or persistence/backdoor deployment."
        }
      ],
      "metrics": [
        {
          "cvssV3_0": {
            "attackComplexity": "HIGH",
            "attackVector": "NETWORK",
            "availabilityImpact": "NONE",
            "baseScore": 8,
            "baseSeverity": "HIGH",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "NONE",
            "scope": "CHANGED",
            "userInteraction": "REQUIRED",
            "vectorString": "CVSS:3.0/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:N",
            "version": "3.0"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-829",
              "description": "CWE-829 Inclusion of Functionality from Untrusted Control Sphere",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-06-03T12:33:10.227Z",
        "orgId": "c09c270a-b464-47c1-9133-acb35b22c19a",
        "shortName": "@huntr_ai"
      },
      "references": [
        {
          "url": "https://huntr.com/bounties/ceb3ce1a-4c45-497a-b25e-cb9a7685e619"
        },
        {
          "url": "https://github.com/huggingface/transformers/commit/676559d5022b74aaa0cee1cee0842b7f27c5320e"
        }
      ],
      "source": {
        "advisory": "ceb3ce1a-4c45-497a-b25e-cb9a7685e619",
        "discovery": "EXTERNAL"
      },
      "title": "Policy Bypass in LightGlue Nested Config Resolution in huggingface/transformers"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "c09c270a-b464-47c1-9133-acb35b22c19a",
    "assignerShortName": "@huntr_ai",
    "cveId": "CVE-2026-5241",
    "datePublished": "2026-06-03T12:33:10.227Z",
    "dateReserved": "2026-03-31T14:26:14.353Z",
    "dateUpdated": "2026-06-03T15:45:36.684Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2",
  "vulnerability-lookup:meta": {
    "epss": {
      "cve": "CVE-2026-5241",
      "date": "2026-06-06",
      "epss": "0.00076",
      "percentile": "0.22894"
    },
    "nvd": "{\"cve\":{\"id\":\"CVE-2026-5241\",\"sourceIdentifier\":\"security@huntr.dev\",\"published\":\"2026-06-03T14:16:46.337\",\"lastModified\":\"2026-06-04T18:54:28.290\",\"vulnStatus\":\"Analyzed\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"A vulnerability in the LightGlue model loading path of huggingface/transformers version 5.2.0 allows an attacker-controlled model repository to execute arbitrary code during model initialization. The issue arises because the `trust_remote_code` parameter, intended to prevent remote code execution, is overridden by untrusted serialized configuration data in a nested code path. Specifically, when loading a LightGlue model using `AutoModel.from_pretrained()` with `trust_remote_code=False`, the `LightGlueConfig` reads the `trust_remote_code` value from the untrusted `config.json` file and propagates it into nested `AutoConfig.from_pretrained()` calls. This results in the execution of attacker-provided Python modules, even when the victim explicitly disables remote code execution. The vulnerability poses a high risk for environments such as API inference servers, research notebooks, CI/CD pipelines, and model evaluation workers, potentially leading to credential theft, lateral movement, or persistence/backdoor deployment.\"}],\"metrics\":{\"cvssMetricV31\":[{\"source\":\"nvd@nist.gov\",\"type\":\"Primary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H\",\"baseScore\":9.6,\"baseSeverity\":\"CRITICAL\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"LOW\",\"privilegesRequired\":\"NONE\",\"userInteraction\":\"REQUIRED\",\"scope\":\"CHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"HIGH\",\"availabilityImpact\":\"HIGH\"},\"exploitabilityScore\":2.8,\"impactScore\":6.0}],\"cvssMetricV30\":[{\"source\":\"security@huntr.dev\",\"type\":\"Secondary\",\"cvssData\":{\"version\":\"3.0\",\"vectorString\":\"CVSS:3.0/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:N\",\"baseScore\":8.0,\"baseSeverity\":\"HIGH\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"HIGH\",\"privilegesRequired\":\"NONE\",\"userInteraction\":\"REQUIRED\",\"scope\":\"CHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"HIGH\",\"availabilityImpact\":\"NONE\"},\"exploitabilityScore\":1.6,\"impactScore\":5.8}]},\"weaknesses\":[{\"source\":\"security@huntr.dev\",\"type\":\"Secondary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-829\"}]}],\"configurations\":[{\"nodes\":[{\"operator\":\"OR\",\"negate\":false,\"cpeMatch\":[{\"vulnerable\":true,\"criteria\":\"cpe:2.3:a:huggingface:transformers:5.2.0:*:*:*:*:*:*:*\",\"matchCriteriaId\":\"B5899C6C-6DEF-448B-9DB5-286D2DB70013\"}]}]}],\"references\":[{\"url\":\"https://github.com/huggingface/transformers/commit/676559d5022b74aaa0cee1cee0842b7f27c5320e\",\"source\":\"security@huntr.dev\",\"tags\":[\"Patch\"]},{\"url\":\"https://huntr.com/bounties/ceb3ce1a-4c45-497a-b25e-cb9a7685e619\",\"source\":\"security@huntr.dev\",\"tags\":[\"Exploit\",\"Third Party Advisory\"]},{\"url\":\"https://huntr.com/bounties/ceb3ce1a-4c45-497a-b25e-cb9a7685e619\",\"source\":\"134c704f-9b21-4f2e-91b3-4a467353bcc0\",\"tags\":[\"Exploit\",\"Third Party Advisory\"]}]}}",
    "vulnrichment": {
      "containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2026-5241\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"poc\"}, {\"Automatable\": \"no\"}, {\"Technical Impact\": \"total\"}], \"version\": \"2.0.3\", \"timestamp\": \"2026-06-03T15:45:14.993177Z\"}}}], \"references\": [{\"url\": \"https://huntr.com/bounties/ceb3ce1a-4c45-497a-b25e-cb9a7685e619\", \"tags\": [\"exploit\"]}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2026-06-03T15:45:33.083Z\"}}], \"cna\": {\"title\": \"Policy Bypass in LightGlue Nested Config Resolution in huggingface/transformers\", \"source\": {\"advisory\": \"ceb3ce1a-4c45-497a-b25e-cb9a7685e619\", \"discovery\": \"EXTERNAL\"}, \"metrics\": [{\"cvssV3_0\": {\"scope\": \"CHANGED\", \"version\": \"3.0\", \"baseScore\": 8, \"attackVector\": \"NETWORK\", \"baseSeverity\": \"HIGH\", \"vectorString\": \"CVSS:3.0/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:H/A:N\", \"integrityImpact\": \"HIGH\", \"userInteraction\": \"REQUIRED\", \"attackComplexity\": \"HIGH\", \"availabilityImpact\": \"NONE\", \"privilegesRequired\": \"NONE\", \"confidentialityImpact\": \"HIGH\"}}], \"affected\": [{\"vendor\": \"huggingface\", \"product\": \"huggingface/transformers\", \"versions\": [{\"status\": \"affected\", \"version\": \"unspecified\", \"lessThan\": \"5.5.0\", \"versionType\": \"custom\"}]}], \"references\": [{\"url\": \"https://huntr.com/bounties/ceb3ce1a-4c45-497a-b25e-cb9a7685e619\"}, {\"url\": \"https://github.com/huggingface/transformers/commit/676559d5022b74aaa0cee1cee0842b7f27c5320e\"}], \"descriptions\": [{\"lang\": \"en\", \"value\": \"A vulnerability in the LightGlue model loading path of huggingface/transformers version 5.2.0 allows an attacker-controlled model repository to execute arbitrary code during model initialization. The issue arises because the `trust_remote_code` parameter, intended to prevent remote code execution, is overridden by untrusted serialized configuration data in a nested code path. Specifically, when loading a LightGlue model using `AutoModel.from_pretrained()` with `trust_remote_code=False`, the `LightGlueConfig` reads the `trust_remote_code` value from the untrusted `config.json` file and propagates it into nested `AutoConfig.from_pretrained()` calls. This results in the execution of attacker-provided Python modules, even when the victim explicitly disables remote code execution. The vulnerability poses a high risk for environments such as API inference servers, research notebooks, CI/CD pipelines, and model evaluation workers, potentially leading to credential theft, lateral movement, or persistence/backdoor deployment.\"}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-829\", \"description\": \"CWE-829 Inclusion of Functionality from Untrusted Control Sphere\"}]}], \"providerMetadata\": {\"orgId\": \"c09c270a-b464-47c1-9133-acb35b22c19a\", \"shortName\": \"@huntr_ai\", \"dateUpdated\": \"2026-06-03T12:33:10.227Z\"}}}",
      "cveMetadata": "{\"cveId\": \"CVE-2026-5241\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2026-06-03T15:45:36.684Z\", \"dateReserved\": \"2026-03-31T14:26:14.353Z\", \"assignerOrgId\": \"c09c270a-b464-47c1-9133-acb35b22c19a\", \"datePublished\": \"2026-06-03T12:33:10.227Z\", \"assignerShortName\": \"@huntr_ai\"}",
      "dataType": "CVE_RECORD",
      "dataVersion": "5.2"
    }
  }
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.





Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

Sightings

Author Source Type Date Other

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or observed by the user.
  • Confirmed: The vulnerability has been validated from an analyst's perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
  • Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
  • Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
  • Not confirmed: The user expressed doubt about the validity of the vulnerability.
  • Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.


Detection rules are retrieved from Rulezet.