Vulnerability-Lookup

GHSA-7HP6-4W63-5G45

Vulnerability from github – Published: 2026-10-27 15:04 – Updated: 2026-10-27 15:04

Summary

LiteLLM: Privilege Escalation to Proxy Admin via Cross-Domain Reuse of the Salt Key

Details

Summary

An authenticated user with internal_user privileges can escalate to proxy_admin and achieve remote code execution. The proxy uses one encryption key for two purposes: sealing secrets at rest and minting session tokens. An attacker exploits this by requesting a new API key with a crafted metadata field containing a forged admin credential as the "secret" value. The proxy encrypts and returns this payload. When this value is then presented as a bearer token, the proxy decrypts it, trusts the forged admin identity, and grants full administrative access, including the ability to execute arbitrary commands via the MCP stdio endpoint.

Affected versions

Versions later than 1.91.0 are exploitable in the default configuration. Versions 1.87.0 through 1.90.x are exploitable only if EXPERIMENTAL_UI_LOGIN=true was explicitly set.

Mitigation

Upgrade to a patched version as soon as possible.

If you cannot upgrade immediately, set EXPERIMENTAL_UI_LOGIN=false. This disables the vulnerable authentication path but also breaks CLI SSO and Claude Code gateway login.

Impact

An internal user can escalate to proxy admin and execute arbitrary commands on the host.

Discovery Credit: Hoa X. Nguyen (OPSWAT Unit 515)

Severity

9.9 (Critical)


                  
                    CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

7.7 (High)


                  
                    CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Show details on source website

JSON

To clipboard

{
  "affected": [
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "litellm"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.91.0"
            },
            {
              "fixed": "1.100.4"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "litellm"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.101.0"
            },
            {
              "fixed": "1.101.3"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "litellm"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.102.0"
            },
            {
              "fixed": "1.102.2"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "litellm"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.103.0"
            },
            {
              "fixed": "1.103.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "PyPI",
        "name": "litellm"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "1.104.0rc1"
            },
            {
              "fixed": "1.104.0rc2"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-114612"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-269",
      "CWE-345",
      "CWE-441"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-10-27T15:04:14Z",
    "nvd_published_at": "2026-10-22T19:17:37Z",
    "severity": "HIGH"
  },
  "details": "### Summary\n\nAn authenticated user with `internal_user` privileges can escalate to `proxy_admin` and achieve remote code execution. The proxy uses one encryption key for two purposes: sealing secrets at rest and minting session tokens. An attacker exploits this by requesting a new API key with a crafted metadata field containing a forged admin credential as the \"secret\" value. The proxy encrypts and returns this payload. When this value is then presented as a bearer token, the proxy decrypts it, trusts the forged admin identity, and grants full administrative access, including the ability to execute arbitrary commands via the MCP stdio endpoint.\n\n### Affected versions\n\nVersions later than 1.91.0 are exploitable in the default configuration. Versions 1.87.0 through 1.90.x are exploitable only if `EXPERIMENTAL_UI_LOGIN=true` was explicitly set.\n\n### Mitigation\n\nUpgrade to a patched version as soon as possible.\n\nIf you cannot upgrade immediately, set `EXPERIMENTAL_UI_LOGIN=false`. This disables the vulnerable authentication path but also breaks CLI SSO and Claude Code gateway login.\n\n### Impact\n\nAn internal user can escalate to proxy admin and execute arbitrary commands on the host.\n\n**Discovery Credit:** Hoa X. Nguyen (OPSWAT Unit 515)",
  "id": "GHSA-7hp6-4w63-5g45",
  "modified": "2026-10-27T15:04:14Z",
  "published": "2026-10-27T15:04:14Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/security/advisories/GHSA-7hp6-4w63-5g45"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-114612"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/commit/1d5fdc87fed66b49e1df8cf7d708ad93249eb7d0"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/commit/9161d7f9d6073603ba61ee4aa5e2c0ef5b37a81b"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/commit/9fa1a641119dd0d4fe43e93622eae5f482ceb63f"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/commit/a80eacacee118fbed54e3d98f239d17a63fac818"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/commit/f7cd90f09e9716652b38daec9f0640cdf49d582b"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/BerriAI/litellm"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/releases/tag/v1.100.4"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/releases/tag/v1.101.3"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/releases/tag/v1.102.2"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/releases/tag/v1.103.1"
    },
    {
      "type": "WEB",
      "url": "https://github.com/BerriAI/litellm/releases/tag/v1.104.0-rc.2"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
      "type": "CVSS_V3"
    },
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N",
      "type": "CVSS_V4"
    }
  ],
  "summary": "LiteLLM: Privilege Escalation to Proxy Admin via Cross-Domain Reuse of the Salt Key"
}

CVE-2026-114612 (GCVE-0-2026-114612)

Vulnerability from cvelistv5 – Published: 2026-10-22 18:01 – Updated: 2026-10-22 18:01

Title

LiteLLM: Privilege Escalation to Proxy Admin via Cross-Domain Reuse of the Salt Key

Summary

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From 1.87.0 until 1.100.4, 1.101.3, 1.102.2, 1.103.1, and 1.104.0-rc.2, the proxy reuses one cryptographic context for at-rest secrets and session tokens, allowing an authenticated internal user to have crafted metadata containing a forged administrator credential encrypted and returned by the API-key flow. Replaying that ciphertext as a bearer token causes the proxy to trust the forged proxy administrator identity, permitting administrative access and arbitrary commands through the MCP stdio endpoint. Versions 1.87.0 through 1.90.x require EXPERIMENTAL_UI_LOGIN to be enabled, while versions after 1.91.0 are exploitable in the default configuration. This issue was fixed in LiteLLM 1.100.4, 1.101.3, 1.102.2, 1.103.1, and 1.104.0-rc.2.

Severity

9.9 (Critical)


                        
                          CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H

CWE

CWE-269 - Improper Privilege Management
CWE-345 - Insufficient Verification of Data Authenticity
CWE-441 - Unintended Proxy or Intermediary ('Confused Deputy')

Assigner

GitHub_M CNA under the mitre root

CNA scorecard C 72/100 over 7867 records in the last 180 days details

References

11 references

URL	Tags
https://github.com/BerriAI/litellm/security/advis…	x_refsource_CONFIRM
https://github.com/BerriAI/litellm/commit/1d5fdc8…	x_refsource_MISC
https://github.com/BerriAI/litellm/commit/9161d7f…	x_refsource_MISC
https://github.com/BerriAI/litellm/commit/9fa1a64…	x_refsource_MISC
https://github.com/BerriAI/litellm/commit/a80eaca…	x_refsource_MISC
https://github.com/BerriAI/litellm/commit/f7cd90f…	x_refsource_MISC
https://github.com/BerriAI/litellm/releases/tag/v…	x_refsource_MISC
https://github.com/BerriAI/litellm/releases/tag/v…	x_refsource_MISC
https://github.com/BerriAI/litellm/releases/tag/v…	x_refsource_MISC
https://github.com/BerriAI/litellm/releases/tag/v…	x_refsource_MISC
https://github.com/BerriAI/litellm/releases/tag/v…	x_refsource_MISC

Impacted products

1 product

Vendor	Product	Version
BerriAI	litellm	Affected: >= 1.87.0, < 1.100.4 Affected: >= 1.101.0, < 1.101.3 Affected: >= 1.102.0, < 1.102.2 Affected: >= 1.103.0, < 1.103.1 Affected: >= 1.104.0-rc.1, < 1.104.0-rc.2

Show details on NVD website

JSON

To clipboard

{
  "containers": {
    "cna": {
      "affected": [
        {
          "product": "litellm",
          "vendor": "BerriAI",
          "versions": [
            {
              "status": "affected",
              "version": "\u003e= 1.87.0, \u003c 1.100.4"
            },
            {
              "status": "affected",
              "version": "\u003e= 1.101.0, \u003c 1.101.3"
            },
            {
              "status": "affected",
              "version": "\u003e= 1.102.0, \u003c 1.102.2"
            },
            {
              "status": "affected",
              "version": "\u003e= 1.103.0, \u003c 1.103.1"
            },
            {
              "status": "affected",
              "version": "\u003e= 1.104.0-rc.1, \u003c 1.104.0-rc.2"
            }
          ]
        }
      ],
      "descriptions": [
        {
          "lang": "en",
          "value": "LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From 1.87.0 until 1.100.4, 1.101.3, 1.102.2, 1.103.1, and 1.104.0-rc.2, the proxy reuses one cryptographic context for at-rest secrets and session tokens, allowing an authenticated internal user to have crafted metadata containing a forged administrator credential encrypted and returned by the API-key flow. Replaying that ciphertext as a bearer token causes the proxy to trust the forged proxy administrator identity, permitting administrative access and arbitrary commands through the MCP stdio endpoint. Versions 1.87.0 through 1.90.x require EXPERIMENTAL_UI_LOGIN to be enabled, while versions after 1.91.0 are exploitable in the default configuration. This issue was fixed in LiteLLM 1.100.4, 1.101.3, 1.102.2, 1.103.1, and 1.104.0-rc.2."
        }
      ],
      "metrics": [
        {
          "cvssV3_1": {
            "attackComplexity": "LOW",
            "attackVector": "NETWORK",
            "availabilityImpact": "HIGH",
            "baseScore": 9.9,
            "baseSeverity": "CRITICAL",
            "confidentialityImpact": "HIGH",
            "integrityImpact": "HIGH",
            "privilegesRequired": "LOW",
            "scope": "CHANGED",
            "userInteraction": "NONE",
            "vectorString": "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H",
            "version": "3.1"
          }
        }
      ],
      "problemTypes": [
        {
          "descriptions": [
            {
              "cweId": "CWE-269",
              "description": "CWE-269: Improper Privilege Management",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-345",
              "description": "CWE-345: Insufficient Verification of Data Authenticity",
              "lang": "en",
              "type": "CWE"
            }
          ]
        },
        {
          "descriptions": [
            {
              "cweId": "CWE-441",
              "description": "CWE-441: Unintended Proxy or Intermediary (\u0027Confused Deputy\u0027)",
              "lang": "en",
              "type": "CWE"
            }
          ]
        }
      ],
      "providerMetadata": {
        "dateUpdated": "2026-10-22T18:01:52.768Z",
        "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
        "shortName": "GitHub_M"
      },
      "references": [
        {
          "name": "https://github.com/BerriAI/litellm/security/advisories/GHSA-7hp6-4w63-5g45",
          "tags": [
            "x_refsource_CONFIRM"
          ],
          "url": "https://github.com/BerriAI/litellm/security/advisories/GHSA-7hp6-4w63-5g45"
        },
        {
          "name": "https://github.com/BerriAI/litellm/commit/1d5fdc87fed66b49e1df8cf7d708ad93249eb7d0",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/commit/1d5fdc87fed66b49e1df8cf7d708ad93249eb7d0"
        },
        {
          "name": "https://github.com/BerriAI/litellm/commit/9161d7f9d6073603ba61ee4aa5e2c0ef5b37a81b",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/commit/9161d7f9d6073603ba61ee4aa5e2c0ef5b37a81b"
        },
        {
          "name": "https://github.com/BerriAI/litellm/commit/9fa1a641119dd0d4fe43e93622eae5f482ceb63f",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/commit/9fa1a641119dd0d4fe43e93622eae5f482ceb63f"
        },
        {
          "name": "https://github.com/BerriAI/litellm/commit/a80eacacee118fbed54e3d98f239d17a63fac818",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/commit/a80eacacee118fbed54e3d98f239d17a63fac818"
        },
        {
          "name": "https://github.com/BerriAI/litellm/commit/f7cd90f09e9716652b38daec9f0640cdf49d582b",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/commit/f7cd90f09e9716652b38daec9f0640cdf49d582b"
        },
        {
          "name": "https://github.com/BerriAI/litellm/releases/tag/v1.100.4",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/releases/tag/v1.100.4"
        },
        {
          "name": "https://github.com/BerriAI/litellm/releases/tag/v1.101.3",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/releases/tag/v1.101.3"
        },
        {
          "name": "https://github.com/BerriAI/litellm/releases/tag/v1.102.2",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/releases/tag/v1.102.2"
        },
        {
          "name": "https://github.com/BerriAI/litellm/releases/tag/v1.103.1",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/releases/tag/v1.103.1"
        },
        {
          "name": "https://github.com/BerriAI/litellm/releases/tag/v1.104.0-rc.2",
          "tags": [
            "x_refsource_MISC"
          ],
          "url": "https://github.com/BerriAI/litellm/releases/tag/v1.104.0-rc.2"
        }
      ],
      "source": {
        "advisory": "GHSA-7hp6-4w63-5g45",
        "discovery": "UNKNOWN"
      },
      "title": "LiteLLM: Privilege Escalation to Proxy Admin via Cross-Domain Reuse of the Salt Key"
    }
  },
  "cveMetadata": {
    "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
    "assignerShortName": "GitHub_M",
    "cveId": "CVE-2026-114612",
    "datePublished": "2026-10-22T18:01:52.768Z",
    "dateReserved": "2026-10-21T20:34:25.599Z",
    "dateUpdated": "2026-10-22T18:01:52.768Z",
    "state": "PUBLISHED"
  },
  "dataType": "CVE_RECORD",
  "dataVersion": "5.2"
}

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.

Vulnerability timeline

Sightings and recorded dates, grouped by day · UTC

Order JSON

Publication & reservation Sightings KEV catalogs Other timestamps

Loading timeline…

The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base model, served locally by ML-Gateway. They have not been verified by an analyst and are provided for guidance only.
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.

Related by attack behaviour

Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.

Action not permitted

GHSA-7HP6-4W63-5G45

Summary

Affected versions

Mitigation

Impact

CVE-2026-114612 (GCVE-0-2026-114612)

Tags

Sightings

Nomenclature

Vulnerability timeline

Related by attack behaviour