ls-consult-polymaths

Multi-perspective expert analysis for an architecture review is generated by spawning multiple polymathic consultant agents in parallel and synthesizing their outputs. This mechanism applies domain-based agent selection to evaluate your task and returns a final synthesis covering consensus, tensions, and recommended actions.

Multi-agent threat modeling works by spawning read-only consultant agents that apply named historical cognitive lenses to analyze your system rather than execute code. It chooses polymaths based on domain signals like adversarial review, combining their parallel outputs into a synthesized outcome of tensions and surprise insights.

Yes, you can use specific historical figures for debugging decision support by naming them directly in your prompt. The system activates targeted triggers when you name specific agents like Feynman or Carmack, applying their cognitive lenses to analyze your debugging task and synthesize recommended actions.

To evaluate product and UX decisions, you simply ask to consult the polymaths with your specific task. The system performs parallel agent orchestration by selecting domain-relevant polymaths, prompting your task verbatim to each agent, and delivering a synthesis of consensus, tensions, and recommended next steps.

The best way to compare expert opinions on technical architecture tradeoffs is through parallel multi-agent consulting. This approach prompts your task verbatim to multiple domain-selected polymath agents and synthesizes the results, explicitly highlighting consensus, disagreements, and surprise insights for a recommended action.

A limitation of using multi-agent analysis for strategy and marketing decisions is that the spawned consultant agents are read-only and perform analysis rather than execution. Additionally, the system's effectiveness depends on matching your domain signals to appropriate polymathic cognitive lenses for accurate decision support.