BerriAI LiteLLM
cpe:2.3:a:litellm:litellm:*:*:*:*:*:*:*
- >= 1.74.2, < 1.83.7
A vulnerability in LiteLLM versions 1.74.2 prior to 1.83.7 allows authenticated users to execute arbitrary commands on the host machine. This issue arises from two MCP test endpoints, 'POST /mcp-rest/test/connection' and 'POST /mcp-rest/test/tools/list', which accepted full server configurations including command execution details. When these endpoints were used with a standard input/output configuration, they executed the specified commands as subprocesses on the proxy host, using the privileges of the proxy process. The vulnerability was accessible to any user with a valid proxy API key, including those with low-privilege internal-user keys.
Exploitation of this vulnerability allows for authenticated users to execute arbitrary commands on the host machine where LiteLLM is running.
Users can upgrade to LiteLLM version 1.83.7 or later, where this vulnerability has been patched. If an immediate upgrade is not possible, the vulnerable endpoints can be blocked at the reverse proxy or API gateway.
Our algorithm analyzes dozens of metrics to generate these 8 key vulnerability categories, which are then combined to calculate the overall risk score.