What problem does it solve?
Before an AI pipeline researches, outlines, or drafts an academic paper, it needs to know the user's writing preferences. This Skill extracts those preferences from the original request and produces a standardized preference brief, so downstream steps never guess or invent requirements.
Core Features & Use Cases
- Dual Decision Modes: Chooses DIRECT mode for immediate generation with conservative academic defaults, or PREFERENCE_DRIVEN mode when the user supplies concrete requirements or requests an interview.
- Structured Output Contract: Emits a fixed PAPER_PREFERENCES brief covering topic, audience, venue style, language, depth, citation style, emphasis, must-includes, avoidances, open questions, and defaults used.
- Non-Blocking Interviews: When the user wants to discuss details first, it lists concise questions while still providing safe defaults so the pipeline can continue.
- Use Case: A user asks for a survey paper on retrieval-augmented generation in author-year citation style; the Skill records those stated preferences exactly and marks all unspecified fields as defaults.
Quick Start
Extract the paper-writing preferences from my request to write a systems paper about vector databases for a practitioner audience.
