Open-source AI proxy LiteLLM compromised in supply-chain attack

On October 21, 2026, the open-source AI proxy LiteLLM was compromised when an attacker injected malicious code into the project. The malicious artifacts were published to PyPI, Docker Hub, and GHCR, according to a statement from LiteLLM on X.

LiteLLM is a widely used open-source proxy that routes application requests to multiple large language model providers. Security researchers said the injected code included an information stealer targeting cloud credentials, passwords, and AI service credentials, as well as a remote access trojan that could allow attackers to execute commands on compromised systems.

LiteLLM said it has removed the malicious releases and container images and advised anyone who installed the affected versions to follow the remediation guidance it published on GitHub.