research

Multi-viewpoint research analyzes complex topics using parallel AI agents to investigate direct, practitioner, contrarian, first principles, and prior art perspectives, synthesizing them into a comprehensive document.

Synthesizing diverse perspectives is best achieved through sub-agent orchestration, leveraging five parallel Sonnet agents for in-depth analysis. This method consolidates findings from multiple analytical viewpoints into a single coherent research document.

You can use either free-text topics or provided documents for analysis. The research process supports both input methods and offers options for adding contextual information and iterative user clarification to refine the investigation.

Multi-agent research works effectively for understanding new and complex technologies by exploring core mechanics, practical usage, potential pitfalls, fundamental problems solved, and historical parallels. This structured analysis builds a comprehensive understanding before implementation planning.

Limitations of using parallel agents for topic exploration include the reliance on Sonnet sub-agents for in-depth analysis, which may require significant processing resources. The structured five-perspective approach is designed for complex subjects and may be excessive for simple inquiries.