Hackers Can Turn Vulnerable LiteLLM AI Gateways Into Root Access and Cloud Credential Theft
LiteLLM is a free tool that lets you run AI chatbots on your own computer. Attackers can break into unsecured LiteLLM setups to steal cloud credentials and even take full control of the server.
- Report priority
- Medium
- Targets
- LiteLLM
How it works
- LiteLLM is a free tool that lets you run AI chatbots on your own computer.
- Attackers can break into unsecured LiteLLM setups to steal cloud credentials and even take full control of the server.
What to do
Check if you have LiteLLM running on a server or cloud with the default password 'sk-1234' or no password at all. If you installed LiteLLM without changing the default password, you are affected. LiteLLM versions older than 1.0.0 are especially risky.
Update LiteLLM to version 1.0.0 or later. Change the default password to a strong, unique one. If you are unsure how to check or update, visit the official LiteLLM documentation or contact your hosting provider for help.
Technical details
An attacker sends a request to an exposed LiteLLM gateway using the default password 'sk-1234' or no password at all. If the gateway is running an older version or has no password set, the attacker can steal cloud credentials and run commands on the server as if they were the owner.
Nearly one in 10 internet-exposed LiteLLM AI gateways accepted the widely documented default master key, sk-1234, or required no authentication, creating a direct path to LLMjacking, sensitive credential exposure, and in vulnerable versions root-level code execution inside the gateway container.