idea-creator

Automating research idea generation and validation requires a multi-phase process involving landscape surveys, LLM brainstorming, mechanical consolidation, and cross-model jury assessment. This Skill systematically generates, validates, and ranks novel research directions using parallel processing.

Validating research feasibility through pilot experiments is handled by executing empirical tests on top-ranked ideas. The process leverages experiment execution and monitoring tools to gather evidence, systematically ranking validated ideas based on actual pilot experiment outcomes.

Cross-model jury assessment can rank brainstormed research ideas by evaluating generated concepts across multiple models. This Skill integrates manual review and codex tools to systematically consolidate and assess idea quality before running pilot experiments.

A research wiki is required to use research automation for idea validation, serving as the foundational knowledge base. The automation process depends on this wiki alongside review and experiment execution tools to generate, validate, and rank novel directions.

Parallel processing improves research idea generation by simultaneously leveraging multiple LLM brainstorming threads across different angles. This approach creates a diverse landscape of concepts that undergo mechanical consolidation and cross-model jury assessment for ranking.

LLM brainstorming for research idea generation is limited by the need for subsequent mechanical consolidation and manual review. Generated ideas require cross-model jury assessment and pilot experiments to verify actual feasibility before proceeding with full research.