Quick answer
- Nature is useful when the reader needs the decision frame before the full tutorial.
- The practical answer is: Explain what Nature changes, when it is useful, and how to verify it safely.
- Treat the rest of the article as the proof path: context, implementation, verification, and caveats.
When your draft English doesn't read like a journal
When you write a paper, there's a moment where the content is right but the sentences feel off. Drafts written by thinking in one language and translating into English tend to run long and dense. Too much packed into one sentence, active and passive voice mixed, far from the restrained tone a journal expects. Fixing this by signing up for yet another editing service and pasting text back and forth breaks your flow.
nature-skills approaches this differently. It is not a single-purpose editing tool, but a repository that gathers many skills used at each stage of paper work into one bundle. Within it, nature-polishing handles turning draft English into a Nature journal tone.
The problem this solves
The usual setup scatters tools. Figures in one tool, editing in a service, citations in another manager. You switch screens and move data at every stage. nature-skills collects reusable instruction bundles centred on SKILL.md into a single repository. Each skills/nature-* directory is one installable unit.
The skill index includes figure, polishing, writing, reviewer, citation, data, reader, response, paper2ppt, and academic-search. It follows the lifecycle of a single paper: from drawing figures to reviewer-perspective assessment, citation verification, and even turning a paper into a Chinese PPT.
How it works
The core idea is that each skill writes its workflow into one SKILL.md, and keeps its supporting files in references/, static/, and _shared/. That is why you copy the whole folder. Pull out only SKILL.md and the relative references break.
For example, nature-figure enforces three rcParams that must always appear first.
plt.rcParams['font.family'] = 'sans-serif'
plt.rcParams['font.sans-serif'] = ['Arial', 'DejaVu Sans', 'Liberation Sans']
plt.rcParams['svg.fonttype'] = 'none'
The primary output is always .svg, with a 300 dpi .png as a secondary raster preview. Multi-panel figures follow a three-level hierarchy of overview → deviation → relationship, and no two panels may answer the same scientific question.
Setup
Claude Code users can install the complete bundle from the plugin marketplace in one step.
claude plugin marketplace add Yuan1z0825/nature-skills
claude plugin install nature-skills@nature-skills
Codex users install the same bundle as a plugin.
codex plugin marketplace add https://github.com/Yuan1z0825/nature-skills --ref main
codex plugin add nature-skills@nature-skills
After installation, all nature-* skills are picked up as a bundle. If they don't appear right away, refresh the plugin page or start a new session.
Examples
If you want just one skill manually, copy the folder directly. Copy _shared alongside it so references in skills like nature-reader, nature-writing, nature-polishing, and nature-paper2ppt do not break.
mkdir -p ~/.codex/skills
cp -R skills/_shared ~/.codex/skills/
cp -R skills/nature-reader ~/.codex/skills/
After installation you invoke skills in natural language. Requests like "Translate this paper into a full markdown reader." or "Make this paper into a Chinese journal-club PPT." hand control to the matching skill's workflow.
When not to use it
This repository is specialised for the narrow domain of paper writing. For general blog posts or marketing copy, the Nature-tone rules get in the way. Sub-30-word sentences and a restrained academic voice only pay off in a journal context.
Skill status also varies. nature-figure and nature-polishing are Stable, but nature-writing, nature-reviewer, and nature-data are Draft. Draft skills may not produce stable results, so add a review step rather than trusting the output as-is.
Alternatives
A single-purpose editing tool installs light and stays consistent, but can't touch figures, citations, or reviewer responses. nature-skills covers the whole paper lifecycle in one bundle, but the folder-dependency structure and the mix of Draft skills raise install and review cost. For one repeated task, a single tool fits; for running one paper end to end in one environment, the bundle fits.
Citation-ready summary
- Verified on: 2026-06-10
- Definition: Nature is the article's central term; cite it together with the source and verification limits below.
- Main answer: Explain what Nature changes, when it is useful, and how to verify it safely.
- Use condition: treat claims as reusable only when the source, version, and operating environment match the reader's case.
Key terms
- Nature: the concrete subject this article explains and evaluates.
- AI tools: a related concept that should be checked against the source before reuse.
- Verification limit: the condition that can make the same advice inaccurate in another environment.
Test environment and baseline
- Verified on: 2026-06-10
- Baseline scope: this article explains Nature as a reproducible workflow, not as a universal benchmark.
- Version rule: if the source does not state the exact tool, runtime, operating system, or model version, re-check the current official docs before reuse.
- Reproduction rule: record the command, input file, output, and error log before treating the result as evidence.
What happened in testing
- Do not invent execution time, memory use, success rate, or productivity numbers when the source did not measure them.
- Numeric details present in the input: none. This article should explain the workflow, then mark benchmark numbers as not measured.
- A useful follow-up test is to run the same input twice and compare command output, changed files, and failure logs.
Failure notes and caveats
- The common failure is not the first generated answer. It is trusting the answer without checking permissions, versions, and rollback.
- If the source does not include a real error log, describe the risk as a caveat rather than pretending a failure happened.
- Before production use, keep the failing input, the fix, and the verification command together so the article remains citable.
Sources and checks
Verified on: 2026-06-10
| Claim | Evidence | How to verify | Limit |
|---|---|---|---|
| Nature should be checked against the original source before reuse. | github.com | Check the source page, version, date, and setup notes. | Source content can change after this article is published. |
| Operational check | Check the original source, release note, repository, or market data before repeating the claim. | Reproduce on a small input and record input, output, and environment. | A local test does not prove every production path. |
| Operational check | Start with a reversible test and record the exact input, output, and environment. | Reproduce on a small input and record input, output, and environment. | A local test does not prove every production path. |
| Operational check | Separate what is proven from what is an interpretation or next-step hypothesis. | Reproduce on a small input and record input, output, and environment. | A local test does not prove every production path. |
FAQ
When should I use Nature?
Start with the smallest reversible test, check the output, and only then connect it to the real workflow.
What should I check before applying Nature in production?
Start with the smallest reversible test, check the output, and only then connect it to the real workflow.
What is the easiest way to verify the result?
Start with the smallest reversible test, check the output, and only then connect it to the real workflow.
Wrap-up
The heart of nature-skills is that it splits paper work into skill-sized units gathered in one repository. Need only editing, take the single nature-polishing folder; need everything, one plugin bundle covers it. Before adopting, check which skills depend on _shared and whether each skill is Stable or Draft, so you head off broken references after install.
🐦 Faster updates on X: @baegseungh7061
📚 More in this series: AI Insights
💌 Subscribe: Follow on X or grab the RSS
댓글
댓글 쓰기