CVE-2026-42271: BerriAI LiteLLM Command Injection Vulnerability

CISA Known Exploited Vulnerabilities

Name

BerriAI LiteLLM Command Injection Vulnerability

Vendor

BerriAI

Product

LiteLLM

Added to the catalog

2026-06-08

Due date for US federal agencies

2026-06-22

Weaknesses

CWE-78, CWE-77

Catalog version

2026.10.30

BerriAI LiteLLM contains a command injection vulnerability that could allow any authenticated user, including holders of low-privilege internal-user keys, to run arbitrary commands on the host.

Required action (CISA)

Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable.

Ransomware use: CISA does not mark known use of this vulnerability in ransomware campaigns.

Links in CISA's notes

• https://github.com/BerriAI/litellm/security/advisories/GHSA-v4p8-mg3p-g94g
• https://github.com/BerriAI/litellm/releases/tag/v1.83.7-stable
• https://nvd.nist.gov/vuln/detail/CVE-2026-42271

Read on 2026-11-02T23:39:54.475Z under licence cisa-kev-cc0@b0067899 from https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json

NVD analysis

Status

Analyzed

Published

2026-05-08T04:16:21.820

Last modified

2026-10-26T20:10:00.160

Weaknesses

CWE-77, CWE-78

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From version 1.74.2 to before version 1.83.7, two endpoints used to preview an MCP server before saving it — POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list — accepted a full server configuration in the request body, including the command, args, and env fields used by the stdio transport. When called with a stdio configuration, the endpoints attempted to connect, which spawned the supplied command as a subprocess on the proxy host with the privileges of the proxy process. The endpoints were gated only by a valid proxy API key, with no role check. Any authenticated user — including holders of low-privilege internal-user keys — could therefore run arbitrary commands on the host. This issue has been patched in version 1.83.7.

References

• https://github.com/BerriAI/litellm/releases/tag/v1.83.7-stable (Product, Release Notes)
• https://github.com/BerriAI/litellm/security/advisories/GHSA-v4p8-mg3p-g94g (Mitigation, Patch, Vendor Advisory)
• https://access.redhat.com/errata/RHSA-2026:27784 (Third Party Advisory)
• https://access.redhat.com/errata/RHSA-2026:28960 (Third Party Advisory)
• https://access.redhat.com/errata/RHSA-2026:30056 (Third Party Advisory)
• https://access.redhat.com/security/cve/CVE-2026-42271 (Third Party Advisory)
• https://bugzilla.redhat.com/show_bug.cgi?id=2467924 (Third Party Advisory)
• https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-42271.json (Third Party Advisory)
• https://www.cisa.gov/known-exploited-vulnerabilities-catalog?field_cve=CVE-2026-42271 (US Government Resource)

This product uses the NVD API but is not endorsed or certified by the NVD.

Read on 2026-11-02T23:42:31.494Z under licence nvd-api-terms-of-use@ff2a5288 from https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-42271

CVE record

Title (CNA)

LiteLLM: Authenticated command execution via MCP stdio test endpoints

Assigner

GitHub_M

State

PUBLISHED

Published

2026-05-08T03:35:16.758Z

Updated

2026-07-15T00:57:49.325Z

Problem types

CWE-77: Improper Neutralization of Special Elements used in a Command ('Command Injection'); CWE-78: Improper Neutralization of Special Elements used in an OS Command ('OS Command Injection')

LiteLLM is a proxy server (AI Gateway) to call LLM APIs in OpenAI (or native) format. From version 1.74.2 to before version 1.83.7, two endpoints used to preview an MCP server before saving it — POST /mcp-rest/test/connection and POST /mcp-rest/test/tools/list — accepted a full server configuration in the request body, including the command, args, and env fields used by the stdio transport. When called with a stdio configuration, the endpoints attempted to connect, which spawned the supplied command as a subprocess on the proxy host with the privileges of the proxy process. The endpoints were gated only by a valid proxy API key, with no role check. Any authenticated user — including holders of low-privilege internal-user keys — could therefore run arbitrary commands on the host. This issue has been patched in version 1.83.7.

References

• https://github.com/BerriAI/litellm/security/advisories/GHSA-v4p8-mg3p-g94g
• https://github.com/BerriAI/litellm/releases/tag/v1.83.7-stable

Read on 2026-11-03T02:20:07.203Z under licence cve-terms-of-use@81bfa2f3 from https://raw.githubusercontent.com/CVEProject/cvelistV5/main/cves/2026/42xxx/CVE-2026-42271.json

GitHub Advisory Database

Also known as / affected packages

Defensive techniques (MITRE D3FEND)

D3FEND maps these defensive techniques to the weakness (CWE) that NVD lists for this CVE, through the digital artifacts the weakness affects. D3FEND does not say they stop this vulnerability.

MITRE D3FEND™ knowledge graph. © 2025 The MITRE Corporation.
"The MITRE Corporation (MITRE) hereby grants you a non-exclusive, royalty-free license to use D3FEND for research, development, and commercial purposes. Any copy you make for such purposes is authorized provided that you reproduce MITRE's copyright designation and this license in any such copy." Terms: https://d3fend.mitre.org/tou/