refining-problem

To refine a research idea, you provide a problem slug and an initial description, and the Skill autonomously maps literature to iteratively define scope, key unknowns, and success criteria. It outputs a standardized problem.md file capturing the research-ready formulation.

Defining research scope and key unknowns requires synthesizing prior work and boundary conditions from existing literature. This Skill launches background research agents to extract definitions and constraints, ensuring your problem formulation is structured and literature-backed.

To assess research question novelty, the Skill synthesizes mapped literature findings and iteratively evaluates problem quality. It sets an immutable novelty requirement based on prior work, determining whether novel research is required for your formulation.

No predefined references or dependencies are required. You only need to supply a problem slug and an initial informal description; the Skill autonomously conducts background literature mapping and generates the structured problem.md file.

The problem.md generation focuses strictly on academic and technical research questions. It relies on autonomous literature mapping to define constraints and success criteria, meaning output quality depends heavily on the clarity of your initial description.

Yes, autonomous agents can structure constraints and success criteria for technical research. The Skill applies iterative refinement and background task agents to transform informal descriptions into research-ready formulations with standardized scope.