tag: Ai-Gateway · 2 items
- Engineer — Act: Active supply chain compromise: if you run litellm, check immediately for version 1.107.0 in pip environments and Docker images, remove it, rotate any credentials and LLM API keys that may have been exposed through the credential stealer.
- SOC/IR — Act: Sweep all systems for litellm==1.107.0 installations deployed on or after October 21, 2026; hunt for outbound connections or credential exfiltration from hosts where that version ran, and treat any match as a potential backdoor compromise.
- Leader — Act: LiteLLM is widely used as an AI gateway routing LLM API keys — confirm whether any teams use it, check exposure to the malicious 1.107.0 release, and request LiteLLM’s incident report to assess scope before this surfaces in board-level AI security discussions.
- Engineer — Act: If you run LiteLLM, immediately check whether your admin key is still “sk-1234” and rotate it to a strong credential; a compromised gateway exposes all upstream model API keys and full prompt/completion history to anyone who finds the instance.
- SOC/IR — Plan: Build a detection rule to flag any LiteLLM API requests authenticating with the literal string “sk-1234”, and sweep existing gateway/proxy logs since initial deployment for unauthorized admin activity.
- Leader — Plan: Direct engineering to inventory all internal and vendor-managed LiteLLM deployments and confirm no default admin credentials are in use; a misconfigured AI gateway exposes both uncapped API spend and the full record of what your applications send to model providers.