LiteLLM says hackers poisoned it with GitHub tokens leaked via Claude Code
LiteLLM reported a security incident in which attackers accessed credentials from the AI gateway's logs, reportedly obtained through Anthropic's Claude coding agent. The hackers leveraged these tokens to distribute a compromised version of LiteLLM's open-source software on Wednesday. An incident report signed by LiteLLM's executives outlined the compromise and its implications.
Key points
- Hackers obtained GitHub tokens from LiteLLM logs
- Claude Code vulnerability exploited for credential access
- Poisoned version published to software repository
- Supply-chain attack on open-source AI gateway
Summary and key points written by AI from the headline and a short excerpt, so they can be incomplete or wrong. The Next Web did the reporting; read the original for the full story.
Read the full story at The Next Web
