Security news, read and ranked every 30 minutes.
Exploited. On CISA's list since 2026-06-08.
CVE-2026-42271 · BerriAI LiteLLM · Published 2026-05-08
8.7 out of 10
CVSS rates it high.
91%
Estimated chance of exploitation in the next 30 days, among the top 1% of all scored flaws (EPSS, 2026-10-29).
Yes
CISA has seen it exploited in real attacks. For US government agencies the fix is deadline passed 130 days ago (2026-06-22).
Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable.
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.
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.
From the NVD.
Explore this week’s weaknesses and all learning topics
Understand the kind of weakness behind this flaw. These resources teach the weakness, rather than reproduce this specific CVE. Follow the vendor's advice above to fix the affected product.
Input changes a command that the server runs through its operating system.
CWE-78 · CISA KEV; NVD record: 0b0ca135-0b70-47e7-9f44-1890c2a1c46c; NVD record: security-advisories@github.com
Reading and practice labs · Free account for labs. Basic HTTP knowledge; some labs use Burp Suite.
CWE-77 · CISA KEV; NVD record: security-advisories@github.com. We have not selected a close learning match for this label.
CWE is a catalogue of weakness types. Labels come from the sources shown and can differ or change. Resource matches reviewed 16 October 2026.
No story in the last 14 days has mentioned it.
Exploited flaws from CISA's KEV catalogue. Chance of exploitation from EPSS by FIRST. Descriptions and scores from the NVD. This product uses the NVD API but is not endorsed or certified by the NVD.