What problem does it solve?
Writing a paper in a field-appropriate narrative and style is hard, time-consuming, and often leads to generic structure or mismatched writing choices; this Skill analyzes top relevant papers and generates a project-specific writing style guide with strict guardrails to keep claims grounded.
Core Features & Use Cases
- Project-aware paper analysis: Scans the repository to understand existing experiments, methods, datasets, metrics, and claims before looking for references.
- Targeted reference paper sourcing: Pulls a small set of latest arXiv papers and highest-cited relevant papers using arXiv and Semantic Scholar strategies.
- Multi-dimensional writing extraction: Analyzes selected papers across story arc, abstract pattern, introduction structure, method presentation, experiment design, and language style.
- Generates a reusable project writing guide: Outputs a new project-level skill at the expected Claude Code and Codex locations, including templates for Abstract, Introduction, Related Work, Method, Experiments, and Conclusion.
- Mandatory guardrails: Enforces anti-AI-speak, anti-overclaim, data grounding, consistency, and honesty rules inside the generated guide so the writing model cannot bypass them.
Quick Start
Run the skill with your research direction, idea, and claims by invoking it as: /paper-style-scout "We propose ... Our method ... We claim ...".
