llm-council

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The script makes runtime POST requests to external LLM endpoints (e.g., https://api.openai.com/v1/chat/completions and https://api.anthropic.com/v1/messages) whose returned completions are injected into subsequent ranking and synthesis prompts, so those external URLs directly control agent prompts and are required for operation.