Threat
llm 0.37.2
First reported · Discovered simonwillison.net
Page published
Earliest dated coverage: 11 Nov 2026 · First observed: 12 Nov 2026 · Latest dated coverage: 11 Nov 2026
Coverage timeline
Single-source advisory — one report is available.
Why it matters
LLM users connecting to third-party OpenAI-compatible endpoints should upgrade to 0.37.2 to prevent unintended disclosure of their OpenAI API keys.
LLM 0.37.2 fixes two credential-leakage paths that could send a user's configured OpenAI API key to a third-party OpenAI-compatible provider. Leakage could occur when the key specified by api_key_name had never been stored, or when a model configured with responses: true was run with -o chat_completions 1. The associated advisories are GHSA-v9h7-jj4g-f862 and GHSA-r639-rr7r-x72w.