Nature Polishing

---
name: nature-polishing
description: >-
  Polish, restructure or translate existing scholarly drafts from files or project
  manuscripts while preserving scientific meaning. Use for 根据这个文件润色，投 ASC,
  论文润色、写得更生动自然、读起来顺一点、减少生硬或模板化表达, and journal-specific
  editing, paragraph logic, terminology, Results compression and LaTeX layout.
  Preserve the draft language unless translation is requested. Drafting missing
  manuscript content belongs to nature-writing; file explanation alone is not editing.
---

# Nature-Style Academic Polishing — Router

This skill is split into two layers:

- A **static layer** under `static/` that holds versioned, reusable content fragments (core principles, paper-type playbooks, per-section guidance, language-specific rules, per-journal style).
- A **dynamic layer** (this file plus `manifest.yaml`) that detects the request's axes and loads only the fragments needed for the current job.

Do not try to apply the polishing logic from memory or from this router. Always load fragments from disk as described below.

## Direct file polishing

For manuscript revision, read `static/core/file-polishing.md` through `always_load`.
It governs scope inference, language preservation, proportionate evidence checking
and verification annotations across all journal routes. Short natural-language
requests are sufficient; no fixed prompt or author questionnaire is required.
Older guide examples asking to confirm or fill a template are planning aids, not
mandatory pauses. Explicit scope restrictions and approval stages still apply.
Read `../nature-shared/core/natural-academic-prose.md` for both ordinary polish
and requests for vivid, natural or less formulaic prose. Keep layout-only work on
its existing specialized route.

## Routing protocol

Follow these five steps every time the skill is invoked.

### 1. Load the manifest and the core layer

Read [manifest.yaml](manifest.yaml). It declares the axes (`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, failure-mode diagnosis, ethics, and output format that apply to every polish 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:

- `paper_type` — research / methods / hypothesis / algorithmic / review. Default: research.
- `section` — abstract / intro / results / discussion / conclusion / title / methods. May be multiple. Infer from the draft and context first; use the direct-file workflow for unresolved scope.
- `language` — en or zh-to-en. Detect from the draft itself.
- `journal` — asc / eaai / isa / kbs / neurocomputing / measurement / ieee-trel / eswa / nature / nat-comms / nat-mach-intell / generic.
  Use the manifest’s aliases and applicability rules; default to `generic`.
  Match named journal targets rather than general topic words. Nature means
  the flagship journal; use each configured subjournal’s own key.

State the detected axis values in one short line to the user before proceeding, so they can correct you cheaply.

For Expert Systems with Applications (`ESWA`, `Expert Syst. Appl.`), use
`journal=eswa` and read `static/fragments/journal/eswa.md`. Its shared
ESWA contract overrides incompatible Nature-specific journal defaults;
retain the normal scientific workflow and evidence safeguards. For layout
jobs that skip prose axes, read `../nature-shared/journal-formats/eswa.md` directly
when ESWA submission requirements affect the layout.

### 3. Load the matching fragments

For every axis except `section`, read the file mapped in the manifest.
Skip section routing for layout-only work or free-floating prose without a section context.

Resolve each requested section before loading its guidance:

1. Look up `journal_sections[journal]` in the manifest. If present, read its
   `profile` and check the stated manuscript/domain applicability. Do not assume
   a journal match alone makes a narrow corpus applicable.
2. If applicable and the section has a mapping, read that mapped file **instead
   of** the default `axes.section.values` fragment and its generic deep section
   reference (for example `references/introduction.md`). The mapped file contains
   both the section plan and detailed journal-specific writing/revision logic.
3. If absent or inapplicable, use the original section fragment and on-demand
   references. Keep the recognized journal and its submission requirements;
   do not substitute another journal's profile.
4. For multiple sections, resolve each separately. For a full manuscript, first
   identify the actual sections/functions, then read the relevant mapped guides.
   Do not require separate Related Work or Discussion headings to use their logic.

Paths in the manifest are relative to this skill directory; links inside a
reference are relative to that reference. Journal adaptations live beside their
original counterparts with a `-<journal>.md` suffix and a matching
`-<journal>-zh.md` Chinese version. Writing owns the shared chapter
content; polishing points to those exact files instead of maintaining a second
copy. When a mapped guide requests a finer template, follow its matching journal-specific links. English is the default reference
language; do not load both languages unless comparison is requested. No files are discovered merely by
being placed in a folder. Apply this resolved section choice to later instructions
in core workflows, task fragments, or reference tables as well. With an active
override, do not additionally load the default deep section guide or Nature-only
corpus/style examples; load general examples only when requested or needed and
compatible with the selected guide. Preserve shared evidence/ethics constraints,
terminology, paper-type guidance, language rules, and the original output format.
The selected guide controls section organization where defaults differ; the
current journal contract controls official requirements, never the corpus.

Do **not** read every fragment in `static/`. Load only what step 2 selected.

### 4. Polish using the loaded material

Apply the loaded fragments in this priority order, matching the `paper type -> section job -> paragraph logic -> claim/evidence/boundary -> sentence polish` rule from `core/failure-modes.md`:

1. Paper-type playbook (architecture, writing order).
2. Resolved section-specific job and failure modes from step 3.
3. Journal-specific framing and constraints.
4. Language-specific sentence and paragraph rules (apply last).
5. Core stance and ethics throughout.

Run `static/core/file-polishing.md` for manuscript revisions. Consult relevant evidence when needed; correct within scope or flag unresolved issues rather than inventing content. Run the shared natural-prose check before returning the revision.

For Results, full-main-text compression, main-versus-SI allocation, or prose
added during revision, load `../nature-shared/core/main-text-discipline.md`
before sentence polishing. Classify each result, retain the shortest sufficient
evidence chain, and require every addition to trigger a deletion or replacement
check across the affected paragraph.

For flagship Nature, Nature Communications, Nature Machine Intelligence, or
another Nature Portfolio title, load the matching shared Nature-style corpus
guidance:

- 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`

Preserve claim escalation, the fast question funnel, Introduction–Results
alignment, discovery-centred abstract compression, evidence-bound local
interpretation, and cross-Results synthesis. These defaults were initially
distilled from published NMI papers; treat them as corpus-derived guidance, not
official policy, and obey the target journal's current rules when they differ.

For any Discussion polish or restructuring job, also load
`../nature-shared/core/discussion-argument-language.md`. Use its function labels
to remove Results replay, repair the movement from specific findings to bounded
implications, calibrate modal and reporting verbs to evidence strength, and make
limitations and future work resolve named claim boundaries. Treat it as general
writing guidance, not journal policy.

### 5. Reach for references only when needed

The files under `references/` are deep references, not defaults. Open them on demand per the `references.on_demand` table in the manifest, for example when the user explicitly asks for phrasebank-style alternatives or a stricter style audit.

When the target is Nature Machine Intelligence and exact limits, availability
sections, conference-extension disclosure or production checks affect the
revision, load
`../nature-shared/journal-formats/nature-machine-intelligence.md`.

When the job is a whole manuscript rather than a passage, or the text has already been through more than one round of editing, also load `../nature-shared/core/consistency-sweep.md`. Polishing passage by passage cannot see accumulated drift: one experimental factor under several names, the same quantity in two units, a metric at two precisions, or a superlative the paper's own table contradicts. Sweep for those before working on sentences, and repeat the sweep until a pass finds nothing new.

**Layout/typesetting (排版) requests are different.** If the user asks to fix
*placement* rather than wording — loose/sparse pages, stranded headings, figures
that don't fill the page or split across pages, "Float too large", multi-panel
arrangement, sparse Supplementary Information — skip the prose axes (paper_type,
section, language, journal) and load `references/latex-layout.md` directly. That
file is self-contained: it carries the diagnosis workflow (render → contact-sheet →
read the log), the float-glue and `[H]`/`\clearpage`/`placeins` patterns, and the
"regenerate wide figures taller at the source" rule. Always compile and visually
inspect rendered pages before and after — never judge layout from the `.tex` alone.

## Why this split

- The static layer is versioned and reviewable. Register journal aliases and the journal fragment in the manifest. For journal-specific chapter logic, additionally register a profile and section paths under `journal_sections`; no further router change is needed for the next journal.
- The dynamic layer keeps each invocation cheap: only the fragments relevant to this draft enter context, instead of the full 1000-line monolith.
- The router itself is short on purpose. Update fragments, not this file, when adding scope.