This vulnerability in BerriAI LiteLLM allows remote code execution through a weakness identified as CWE-420, involving an unprotected alternate channel. Specifically, attackers can perform bytecode rewriting by accessing the /guardrails/test_custom_code URI, enabling arbitrary code execution remotely. The CVSS 3.1 score is 8.8, indicating high severity with network attack vector, low attack complexity, requiring low privileges and no user interaction, and impacts confidentiality, integrity, and availability.

Successful exploitation of this vulnerability could allow remote attackers to execute arbitrary code on the affected system, potentially leading to full compromise of the LiteLLM instance. This impacts confidentiality, integrity, and availability of the system. No known exploits are currently reported in the wild.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Since no official fix or remediation level is provided, users should monitor BerriAI's advisories for updates. Until a patch is available, restricting access to the /guardrails/test_custom_code URI and applying network-level controls to limit exposure may reduce risk.