council

To achieve multi-model consensus for strategic decision-making, you query Claude Opus, Gemini, and GPT-5.4 in parallel with the same question. The tool then synthesizes their independent responses into a single unified recommendation, highlighting any disagreements to provide a robust conclusion.

To log LLM advisory sessions for auditability, the tool automatically creates a dated session directory under docs/council-sessions. It saves the original question and each model's reply as separate files, generating a concise synthesis file to ensure a persistent, auditable record of the parallel inquiry.

The best way to compare risk assessments from different AI models is using a parallel tri-model inquiry. By sending the same policy or risk question to Claude Opus, Gemini, and GPT-5.4 simultaneously, you can directly compare their distinct viewpoints and extract a synthesized consensus.

Yes, you can use parallel AI model querying for product and design considerations. The orchestration scope explicitly covers strategic decision-making, risk assessment, and policy guidance across product, business, and design contexts, aggregating diverse advice into a single synthesized view.

No, you do not need to install external dependencies to run tri-LLM consensus. The Skill operates without any listed dependencies, requiring only a prompt to trigger the orchestration of Claude Opus, Gemini, and GPT-5.4, followed by the automatic generation of session logs.