litellm

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes literal API keys and credentials in code/config examples (e.g., api_key="sk-1234", master_key: sk-1234, postgresql://user:pass...) and shows them being embedded directly into client calls and configs, which requires the agent to handle/output secret values verbatim.