Nearly 1 in 10 Exposed LiteLLM Gateways Accepted the Example "sk-1234" Admin Key
LiteLLM, an open-source AI gateway, has a significant security vulnerability stemming from a default administrator key (sk-1234) that allows attackers to access API keys, cloud IAM credentials, and potentially execute code within the gateway container. While the initial vulnerability has been patched, a separate flaw allows unauthenticated access to MCP sessions using single-character tokens. Microsoft has observed active exploitation of these flaws, with attackers leveraging them to steal credentials and install cryptocurrency miners. The vulnerability highlights the importance of treating AI gateways as Tier-0 secrets stores and implementing robust network restrictions and IAM controls.
Summary written automatically in our own words from the original article, which belongs to its publisher and remains the reference. It may contain errors. Sources & data
