Summary

This repository provides Claude "skills" for generating academic content aligned with Nature journal standards. Targeting researchers, it automates scientific figure creation and manuscript prose polishing to streamline publication.

How It Works

Skills derive rules from primary sources (Nature papers, guidelines), not intuition, and are output-first and extensible. nature-figure produces multi-panel matplotlib figures with specific hierarchy, typography, and color palettes, outputting editable SVGs. nature-polishing transforms draft text (including Chinese-English) into prose matching Nature's strict conventions for sentence length, tense, hedging, and citation.

Quick Start & Requirements

Installation commands and explicit requirements are not detailed. A contribution guide outlines skill addition structure (SKILL.md, README.md). No quick-start or demo links are present.

Highlighted Details

• nature-figure supports 10 chart families, emphasizing non-redundant panel architecture and SVG output.
• nature-polishing enforces rules like sentences ≤ 30 words, calibrated hedging, section-aware tense, and British English.
• nature-data (Draft) automates Data Availability statements, repository plans, and FAIR checks, with bilingual support.
• All skills follow principles: primary source grounding, explicit rationales, section-sensitivity, and immediate, usable output.

Maintenance & Community

New skills are under development (e.g., nature-stats, nature-response). The project's collaborative group is full; contact the administrator to join. No specific community channels are listed.

Licensing & Compatibility

No software license is specified. Commercial use compatibility is undetermined.

Limitations & Caveats

The nature-data skill is in "Draft" status. Several key skills (nature-stats, nature-response, nature-methods, etc.) are "Candidate" and not yet built. The collaborative group is at capacity.