CtrlK
BlogDocsLog inGet started
Tessl Logo

nature-polishing

Polish, restructure, or translate academic prose into concise Nature-leaning English while preserving facts, evidence boundaries, terminology, and citation intent. Use for manuscript paragraphs, abstracts, introductions, Results, discussions, conclusions, titles, Methods, Chinese drafts, proofreading, language editing, and general academic or scientific writing. Also use to shorten bloated Results, allocate evidence across main text, captions, and Supplementary Information, prevent reviewer-driven revision accretion, reduce repeated statistics or claims, and apply paragraph-necessity checks. Covers LaTeX layout or typesetting fixes such as sparse pages, stranded headings, oversized or split figures, float errors, multi-panel arrangement, and sparse Supplementary Information via references/latex-layout.md. Trigger on 学术写作、科研写作、论文润色、SCI写作、英文论文润色、语言润色、润色、改写、学术英语、排版.

64

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/nature-polishing/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

60%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The router is well-written as prose — concise, clearly sequenced, with real validation checkpoints — but its progressive-disclosure design is not actually delivered: the manifest and static/nature-shared fragments it routes to are absent from the bundle, and most real reference files are orphaned. The skill as shipped cannot execute its own routing protocol.

Suggestions

Ship manifest.yaml (and the static/ and ../nature-shared/ fragments it maps to) or rewrite the body so it only routes to files that actually exist in the bundle.

Name and link the six present references/ files (writing-strategy.md, section-moves.md, style-guardrails.md, phrasebank-playbook.md, published-article-patterns.md, nat-comms-2025-diction.md) explicitly instead of referring to them only generically as 'files under references/'.

Add a fallback for when manifest.yaml or a mapped fragment is missing, so the routing protocol degrades gracefully instead of failing at step 1.

DimensionReasoningScore

Conciseness

The router is efficient and assumes Claude's competence — e.g. 'Do not try to apply the polishing logic from memory... Always load fragments from disk' — but a few illustrative passages like the drift examples ('one experimental factor under several names, the same quantity in two units...') and the NMI-provenance note could be trimmed, so it sits just below the lean 5 anchor.

4 / 5

Actionability

The 5-step protocol is concretely worded with named files and a clear priority order, but the executable targets are mostly missing — 'Read [manifest.yaml](manifest.yaml)', 'every file listed under always_load', and 'the file mapped in the manifest' all reference artifacts absent from the bundle, leaving the guidance concrete in form but incomplete in practice.

3 / 5

Workflow Clarity

A clear numbered sequence (load manifest/core → detect axes → load fragments → polish in priority order → reach for references) with explicit checkpoints — 'State the detected axis values in one short line to the user', 'Always compile and visually inspect rendered pages before and after', and the 'repeat the sweep until a pass finds nothing new' feedback loop — though the workflow's foundation rests on the absent manifest, a minor gap.

4 / 5

Progressive Disclosure

Scored against the actual bundle: the navigation hub 'manifest.yaml' the router depends on is missing, as are 'static/', 'core/failure-modes.md', and all six '../nature-shared/...' paths; only 'references/latex-layout.md' is both real and linked, while six other references/ files exist but are never named in the body — a minimal/broken structure with buried or dangling references.

2 / 5

Total

13

/

20

Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, comprehensive description that clearly states capabilities, triggers, and scope in both English and Chinese with concrete actions throughout. The only soft spot is mild breadth into generic proofreading/language-editing that slightly raises overlap risk with generic writing skills.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'Polish, restructure, or translate academic prose', 'shorten bloated Results, allocate evidence across main text, captions, and Supplementary Information', 'reduce repeated statistics or claims', and LaTeX fixes like 'sparse pages, stranded headings, oversized or split figures, float errors' — giving comprehensive coverage rather than the minor gaps of a 4.

5 / 5

Completeness

Explicitly answers both what ('Polish, restructure, or translate academic prose into concise Nature-leaning English...') and when ('Use for manuscript paragraphs...', 'Also use to shorten bloated Results...', 'Trigger on...') with concrete trigger phrases, fitting the top anchor exactly.

5 / 5

Trigger Term Quality

Comprehensive natural-term coverage in both languages — '润色, 改写, 学术写作, SCI写作, 英文论文润色, 语言润色, 排版' plus English triggers like 'manuscript paragraphs, abstracts, proofreading, language editing' — matching the anchor's synonym-rich coverage.

5 / 5

Distinctiveness Conflict Risk

The Nature-leaning academic-polishing niche plus Chinese SCI triggers is mostly distinct, but the broadening to 'general academic or scientific writing' and 'proofreading, language editing' creates minor overlap risk with generic writing/proofreading skills, keeping it just below the minimal-conflict anchor 5.

4 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository
Yuan1z0825/nature-skills
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.