CVE-2026-42271
BerriAI LiteLLM Command Injection Vulnerability
This vulnerability is in the CISA Known Exploited Vulnerabilities catalog: attackers are using it. If you run LiteLLM, fix it now.
CISA required action: Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable.
US federal civilian agencies must remediate by Jun 22, 2026.
Patch status
Official fix available since
Apply the vendor's update to every affected system. Check the fixed-in versions below where known.
Basis: NVD patch reference
Description
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.
Source: NIST National Vulnerability Database. Official vulnerability data, reproduced as published.
Affected products
| Product | Vendor | Source |
|---|---|---|
| LiteLLM | BerriAI | KEV |
Sources: KEV = CISA Known Exploited Vulnerabilities catalog. Version details have not been published yet.
Intelligence timeline
Developments from the last 30 days, newest first.
No developments recorded for CVE-2026-42271 in the last 30 days.
1 earlier event is available with a subscription. See plans.
Coverage
We have not published a story about this vulnerability yet.