CVE-2026-42271
BerriAI LiteLLM: Command Injection
As of , CVE-2026-42271 in BerriAI LiteLLM is on CISA's list of exploited vulnerabilities: CISA listed it on . Our patch-first group: Patch this week.
- Exploited
- Yes: CISA listed it on 8 June 2026
- US federal deadline
- 22 June 202614 days after CISA listed it. US federal civilian agencies must fix it by then.
- Used in ransomware campaigns
- Not known to CISA
- EPSS score
- 0.93Higher than 99% of the CVEs EPSS scores. EPSS estimates the chance of exploitation activity in the next 30 days.
- Public exploit
- 1 Metasploit module
- Fix
- Vendor advice: github.comLinks below, from CISA's entry.
What CISA says to do
Apply mitigations per vendor instructions, follow applicable BOD 22-01 guidance for cloud services, or discontinue use of the product if mitigations are unavailable.
CISA's required action
- github.comThis vulnerability affects a common open-source component, third-party library, or a protocol used by different products. Please check…
- github.com
What the flaw is
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.
CISA's 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.
The CVE record's description, from GitHub_M
- CVE published
- 8 May 2026
- Assigned by
- GitHub_M
- CVSS
- 8.7 High (CVSS 4.0, from the CNA)
- 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')
- CISA SSVC, exploitation
- active
- CISA SSVC, automatable
- no
- CISA SSVC, technical impact
- total
Timeline
- The CVE record was published.
- CISA added it to its list of exploited vulnerabilities.
- The US federal deadline to fix it.
Public exploits
Facts from Metasploit's module list and Exploit-DB's index; we never copy exploit code.
- Metasploit: BerriAI LiteLLM Proxy MCP Test Endpoint Command Executionexploit module, rank excellent
LiteLLM: other exploited entries
| Vulnerability | Product | Our group | Listed | Deadline | EPSS |
|---|---|---|---|---|---|
| CVE-2026-42208SQL Injection | BerriAI LiteLLM | Patch this weekMetasploit module | 0.06 | ||
| CVE-2026-59822Improper Authentication | BerriAI LiteLLM | Patch soon | 0.01 |
Read further
- CVE recordcve.org
- NVD entrynvd.nist.gov
- CISA's catalogcisa.gov
- EPSS for this CVEapi.first.org