CVE-2026-37004
Published:August 27, 2026
Updated:September 24, 2026
BerriAI litellm <=1.82.4 is vulnerable to Server-Side Template Injection (SSTI), which allows unauthenticated remote attackers to execute arbitrary OS commands via a crafted dotprompt_content parameter in the /prompts/test endpoint due to use of an unsandboxed jinja2.Environment.
Affected Packages
litellm (CONDA):
Affected version(s) >=1.34.0 <1.83.7Fix Suggestion:
Update to version 1.83.7https://github.com/BerriAI/litellm.git (GITHUB):
Affected version(s) >=v1.74.14-nightly <v1.83.7-stableFix Suggestion:
Update to version v1.83.7-stablelitellm (PYTHON):
Affected version(s) >=0.1.0 <1.83.7Fix Suggestion:
Update to version 1.83.7litellm (PYTHON):
Affected version(s) >=1.74.14 <1.83.7Fix Suggestion:
Update to version 1.83.7Related Resources (6)
Do you need more information?
Contact UsCVSS v4
Base Score:
9.3
Attack Vector
NETWORK
Attack Complexity
LOW
Attack Requirements
NONE
Privileges Required
NONE
User Interaction
NONE
Vulnerable System Confidentiality
HIGH
Vulnerable System Integrity
HIGH
Vulnerable System Availability
HIGH
Subsequent System Confidentiality
NONE
Subsequent System Integrity
NONE
Subsequent System Availability
NONE
CVSS v3
Base Score:
9.8
Attack Vector
NETWORK
Attack Complexity
LOW
Privileges Required
NONE
User Interaction
NONE
Scope
UNCHANGED
Confidentiality
HIGH
Integrity
HIGH
Availability
HIGH
Weakness Type (CWE)
Improper Neutralization of Special Elements Used in a Template Engine
EPSS
Base Score:
0.55