LiteLLM

LiteLLM is an open-source library that provides a unified interface for large language models and AI services. The vulnerability requires network access and user authentication to exploit, but does not require special conditions to trigger; it could allow an authenticated attacker to view sensitive information through a phishing-style attack (indicated by the UI requirement). Security teams should monitor for this issue and plan patching within the vendor's deadline of August 8, 2025, particularly for environments where LiteLLM handles sensitive data or integrates with critical applications.