nature-writing — Claude skill: draft scientific manuscripts
A skill that drafts and restructures scientific manuscript sections and first-submission materials from evidence you supply.
npx skills add Yuan1z0825/nature-skills --skill nature-writing -a claude-codeWhat nature-writing does
nature-writing is a skill for scientific writing in Nature style. It takes the evidence you provide and uses it to draft or restructure manuscript sections: abstract, intro, related work, method, experiments, discussion, conclusion, title. It also assembles a first-submission package.
It works as a router. First it reads manifest.yaml, which declares the axes (task, paper type, section, language, journal) and the files each value maps to. Then it detects the axis values from your request, states them in one line, and loads only the matching fragments. Language is en or zh-to-en. Paper type is research, methods, hypothesis, algorithmic or review, defaulting to research. Journal is nature, nature-family, nat-comms, nat-mach-intell or generic, defaulting to generic.
Results and Discussion pull in separate files on main-text discipline and argument language. If evidence or a boundary is missing, the skill writes a placeholder and lists it under Assumptions or missing inputs: instead of inventing content. Worth installing if you write for Nature Portfolio and keep rebuilding the argument by hand across sections. For language-only edits to finished prose use nature-polishing; for post-decision correspondence use nature-response.
What it helps with
- draft a section from the author's claims and data
- restructure the argument chain in a manuscript
- compress the main text into a short evidence chain
- prepare first-submission materials for a journal
- sort results across main text, captions and SI
Example prompts for the agent
- Draft the Introduction from these results, target journal Nature Communications
- Restructure the Discussion: main finding first, then limitations
- Build the first-submission package, paper type methods
How to install nature-writing
1Skills CLI (Vercel)
npx skills add Yuan1z0825/nature-skills --skill nature-writing -a claude-code2Manually: put the files in place
~/.claude/skills/nature-writing/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/nature-writing/
1Skills CLI (Vercel)
npx skills add Yuan1z0825/nature-skills --skill nature-writing -a codex2Manually: put the files in place
.agents/skills/nature-writing/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add Yuan1z0825/nature-skills --skill nature-writing -a cursor2Manually: put the files in place
.cursor/skills/nature-writing/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Yuan1z0825/nature-skills nature-writing2Manually: put the files in place
.github/skills/nature-writing/SKILL.md1Skills CLI (Vercel)
npx skills add Yuan1z0825/nature-skills --skill nature-writing -a gemini-cli2Platform CLI
gemini skills install https://github.com/Yuan1z0825/nature-skills --path skills/nature-writingManually: put the files in place
.devin/skills/nature-writing/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add Yuan1z0825/nature-skills --skill nature-writing -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
From the same set17
The whole set →Similar
Questions about nature-writing
How do I install nature-writing in Claude?
Run npx skills add Yuan1z0825/nature-skills --skill nature-writing -a claude-code. The skill lands in ~/.claude/skills/nature-writing/SKILL.md.
What does nature-writing do?
It drafts and restructures scientific manuscript sections and first-submission materials from evidence you provide. It routes your request through axes and loads only the fragments that match.
How is nature-writing different from nature-polishing?
nature-writing handles drafting and argument restructuring, while nature-polishing only edits language in existing prose. Post-decision correspondence goes to nature-response.
Source text
SKILL.md · Apache-2.0
--- name: nature-writing description: Draft or restructure scientific manuscript arguments, sections, and initial-submission materials from author-provided evidence. Use for 论文写作、章节起草、论证重构、正文压缩、首次投稿材料. Use nature-polishing for language-only edits to existing prose and nature-response for post-decision correspondence. --- # Nature-Style Scientific Writing — Router ## Routing protocol For a new drafting task, follow the routing below. For follow-up edits, reuse established task choices and already loaded guidance; read additional fragments only when the requested scope changes. ### 1. Load the manifest and the core layer Read [manifest.yaml](manifest.yaml). It declares the axes (`task`, `paper_type`, `section`, `language`, `journal`), the allowed values, and the file paths each value maps to. Also read every file listed under `always_load`. These hold the default stance, writing workflow, and output format that apply to every drafting job. ### 2. Detect the axis values for this request For each axis in the manifest, decide the value using the manifest's `detect:` hint and the user's input: - `task` — manuscript / submission-package. Use `submission-package` for first-submission materials, never for revision correspondence. - `paper_type` — research / methods / hypothesis / algorithmic / review. Default: research. - `section` — abstract / intro / related-work / method / experiments / discussion / conclusion / title. May be multiple. Ask the user if it is ambiguous and matters for the draft. - `language` — en or zh-to-en. Detect from the user's notes themselves. - `journal` — nature / nature-family / nat-comms / nat-mach-intell / generic. Default: generic. Use `nature` only for the flagship journal Nature, `nat-comms` for Nature Communications, `nat-mach-intell` for Nature Machine Intelligence (NMI), and `nature-family` for other Nature Portfolio titles or an unspecified Nature-family request. State the detected axis values in one short line to the user before drafting, so they can correct you cheaply. This is a progress update, not an approval gate; continue unless a necessary decision remains unresolved. ### 3. Load the matching fragments For each axis value, Read the file mapped in the manifest. Skip the `section` axis when the task is `submission-package` or when the user explicitly asks for a free-floating argument paragraph with no section context. Do **not** read every fragment in `static/`. Load only what step 2 selected. ### 4. Draft using the loaded material Apply the loaded fragments in this priority order: 1. Core stance + intake (`core/stance.md`) — surface missing claim / evidence / boundary before drafting. 2. Paper-type playbook — argument chain, drafting order. 3. Section-specific drafting rules and structure. 4. Task-specific submission rules when `task=submission-package`. 5. Journal-specific framing and constraints. 6. Language-specific sentence and paragraph rules (apply last). For `task=manuscript`, use `core/workflow.md` at the requested scale. Plan the argument for a new section or substantial restructuring; a title, single paragraph, or local follow-up needs only the applicable evidence, wording, and consistency checks. Complete the requested prose unless a material unresolved decision blocks it or the user requested an outline for approval first. When drafting or restructuring Results, or compressing a full manuscript's main text, also load `../nature-shared/core/main-text-discipline.md` before building the paragraph map. Classify every result by function, allocate it across main text, captions, Methods/source data, and SI, then draft the shortest sufficient evidence chain. Do not equate a complete analysis record with a complete main text. When the target is flagship Nature, Nature Communications, Nature Machine Intelligence, or another Nature Portfolio title, load the matching shared Nature-style corpus guidance for the section being drafted: - Results or Discussion → `../nature-shared/core/nature-results-discussion.md` - Introduction or whole-manuscript narrative → `../nature-shared/core/nature-introduction.md` - Abstract → `../nature-shared/core/nature-abstract.md` Use these files for claim escalation, question-chain alignment, discovery-centred compression, and synthesis. They were initially distilled from published NMI papers and generalized as Nature-style defaults; do not present them as official policy, and let the target journal's current rules override them. For any Discussion drafting, restructuring, or section audit, also load `../nature-shared/core/discussion-argument-language.md`. Use it to select the opening anchor, control the reverse-funnel expansion, distinguish literature positioning from citation decoration, calibrate modal strength to evidence, and turn limitations and future work into claim-specific reasoning. This is general writing guidance rather than an official journal rule. For `task=submission-package`, follow `static/fragments/task/submission-package.md` and `references/submission-package.md` instead. Build the deliverable matrix and readiness audit; do not force manuscript paragraph architecture onto administrative submission materials. If essential evidence or boundary is missing, write a placeholder and list it under `Assumptions or missing inputs:` instead of inventing content. ### 5. Reach for references only when needed The files under `references/` are deep references and the example library, not defaults. Open them on demand per the `references.on_demand` table in the manifest. Typical triggers: - The user asks for a concrete example or template → `references/examples/index.md`. - A section's draft has structural problems that the section fragment alone does not explain → the matching `references/<section>.md`. - The user needs a broad-audience `Nature` abstract opening or asks about a `summary paragraph` → `references/nature-summary-paragraph.md`. - The user asks "does this paragraph flow?" → `references/paragraph-flow.md`. - The user asks for a self-review or rejection-risk audit → `references/paper-review.md`. - The user asks what belongs in the main text, captions, or SI; wants a shorter Results section; or is adding reviewer-driven explanation → `../nature-shared/core/main-text-discipline.md`. - The user requests a complete first-submission package, templates, or a submission-readiness audit → `references/submission-package.md`. - The target is the flagship journal Nature and exact submission or formatting requirements matter → `../nature-shared/journal-formats/nature.md`. - The target is Nature Machine Intelligence and exact content-type, submission, data/code or production requirements matter → `../nature-shared/journal-formats/nature-machine-intelligence.md`. - Any Nature / Nature Portfolio target needs Results claim progression, evidence-bound interpretation, robustness placement, or Discussion synthesis → `../nature-shared/core/nature-results-discussion.md`. - Any target needs a Discussion function chain, evidence-calibrated modal language, claim-specific limitations, non-redundant literature positioning, or uncertainty-driven future work → `../nature-shared/core/discussion-argument-language.md`. - Any Nature / Nature Portfolio target needs an Introduction funnel, exact gap, literature logic, question-first novelty, study roadmap, or alignment with Results → `../nature-shared/core/nature-introduction.md`. - Any Nature / Nature Portfolio target needs abstract evidence-chain, main/supporting-claim, numeric-result, or final-payoff decisions → `../nature-shared/core/nature-abstract.md`. - The work involves regulated or specialist research compliance → `../nature-shared/core/research-compliance.md`. ## Submission boundary - `nature-writing` owns **initial submission** materials prepared before peer review. - `nature-response` owns revision cover letters, rebuttals, point-by-point responses, marked manuscripts, appeals, and other post-decision correspondence. - Route graphical abstracts and TOC graphics to `nature-figure`; route simulated pre-submission peer review to `nature-reviewer`.
© Yuan1z0825 · Apache-2.0 · Yuan1z0825/nature-skills