What problem does it solve? Turning a vague topic or area of interest into strong, novel, and feasible research questions is hard: questions are often too broad to answer, too narrow to matter, or already settled in the literature. This Skill runs a structured sharpen-score-revise loop that drafts candidate questions, grades each against a fixed rubric, checks novelty against the literature or web, and iteratively rewrites the weakest axis until enough questions clear the bar. ## Core Features & Use Cases - Rubric-based scoring: Each candidate is graded 0-5 on five fixed axes (Specific, Answerable, Novel, Feasible, Significant) by a fresh independent Grader, converted to a 0-100 score with a configurable pass threshold. - Novelty verification: Optional literature search via a sibling literature-search skill, web search, or an explicitly labeled unverified mode, so novelty claims are checked rather than assumed. - Iterative revision loop: Each iteration diagnoses a question's weakest axis and makes one focused revision, logging progress to a ledger until the target number of strong questions is reached or the budget runs out. - Use Case: A graduate student interested in sleep and memory runs the loop and receives three polished questions, each with rubric scores, the closest published work, the study design that would answer it, and why it matters. ## Quick Start Ask the assistant to use the research-question skill to turn your topic of interest into three strong, novel, and feasible research questions.
