CtrlK
BlogDocsLog inGet started
Tessl Logo

scholarly-writing-refiner

Polishes academic English paragraph by paragraph, reviewing grammar, word choice, voice, coherence, and sentence structure. Outputs revision suggestions alongside polished text. Triggered by phrases like 'polish this paragraph,' 'check the grammar,' 'rewrite in academic English,' or keywords like manuscript editing, SCI polishing, and journal submission editing.

72

Quality

88%

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

SKILL.md
Quality
Evals
Security

Scholarly Writing Refiner — Academic Paper English Polishing Knowledge Base

Helps users review and polish English papers paragraph by paragraph according to international academic journal standards. Covers grammar correction, academic word choice optimization, voice normalization, coherence strengthening, sentence variety, and more. Outputs revision suggestions along with the polished text.

Quick Start

Users only need to provide:

  1. Text to polish: One or more paragraphs of English paper content
  2. Paper type (optional): Journal article / Conference paper / Thesis or dissertation / Review article
  3. Target journal/field (optional): e.g., Nature, IEEE, AAAI, Medicine, Computer Science, etc.
  4. Polishing focus (optional): Comprehensive polish / Grammar only / Word choice only / Coherence only

Example:

"Polish this Introduction for me. The target venue is NeurIPS, and I'd like the language to sound more natural with better logical flow."


1. Review Dimensions Overview

Polishing is carried out across 5 dimensions. Each dimension is rated independently with specific revision suggestions:

DimensionLabelReview Focus
GrammarGrammarSubject-verb agreement, tense, articles, prepositions, clause structure, punctuation
Word ChoiceWord ChoiceAcademic register, precision, collocation, avoidance of colloquialisms
VoiceVoice & TenseActive/passive voice selection, tense consistency
CoherenceCoherence & CohesionIntra-paragraph and inter-paragraph transitions, argumentation chain, use of signaling words
Sentence StructureSentence StructureSentence variety, balance of long and short sentences, coordination and subordination

2. Grammar Review Rules

2.1 Common Grammar Errors Checklist

Error TypeIncorrect ExampleCorrectionExplanation
Subject-verb disagreementThe results of the experiment shows...The results of the experiment show...Subject is "results" (plural)
Missing/misused articleWe propose method to solve...We propose a method to solve...Singular countable nouns require an article
Dangling modifierUsing the proposed method, the accuracy was improved.Using the proposed method, we improved the accuracy.The implied subject of a participial phrase must match the main clause subject
Run-on sentenceThe model performs well , it achieves 95% accuracy.The model performs well**;** it achieves 95% accuracy. / The model performs well**. It** achieves 95% accuracy.A comma cannot join two independent clauses
Incomplete comparisonOur method is more efficient.Our method is more efficient than the baseline.Comparatives require an explicit object of comparison
Broken parallelismThe system can detect, classify, and is able to segment...The system can detect, classify, and segment...Coordinated elements must share the same grammatical form
that/which confusionThe model which we proposed...The model that we proposed...Restrictive relative clauses use "that"
Irregular pluralsThese phenomenon indicate...These phenomena indicate...Watch for irregular plural forms

2.2 Punctuation Rules

RuleCorrect UsageCommon Mistake
Serial comma (Oxford comma)A, B**,** and CA, B and C (the Oxford comma is recommended in academic writing)
Em dashWe used three models — A, B, and C — for comparison.An em dash with spaces on both sides, or without spaces (depends on journal style)
Capitalization after colonCapitalize if a complete sentence follows: The result is clear: The model outperforms...Do not capitalize if a fragment follows
Quotation marks and periodsAmerican style: period inside quotes. British style: period outside quotes.Choose based on the target journal's regional convention
Abbreviation periodse.g., i.e., et al., etc.Note the comma: e.g., / i.e.,

2.3 Tense Guidelines by Paper Section

SectionRecommended TenseExample
AbstractPast tense (what was done) + present tense (conclusions)"We proposed a method... The results show that..."
IntroductionPresent tense (current knowledge/consensus) + past tense (prior work)"Deep learning has become... Smith et al. demonstrated that..."
MethodsPast tense (experimental procedures)"We trained the model on... The data were preprocessed..."
ResultsPast tense (experimental findings)"The model achieved 95% accuracy. Table 2 shows..."
DiscussionPresent tense (interpreting significance) + past tense (citing results)"This result suggests that... Our findings indicated that..."
ConclusionPast tense (summarizing work) + present tense (contributions/significance)"We proposed and evaluated... This work contributes to..."

3. Word Choice Optimization Rules

3.1 Colloquial → Academic Substitution Table

ColloquialAcademic AlternativeContext Notes
a lot ofnumerous / a substantial number of / considerableChoose based on what is being modified
getobtain / acquire / achieve / attainChoose based on collocation
showdemonstrate / illustrate / indicate / reveal"demonstrate" emphasizes proof; "indicate" emphasizes suggestion
big / hugesubstantial / significant / considerable
thingfactor / aspect / element / component
goodeffective / favorable / advantageous / robust
badadverse / detrimental / suboptimal / inferior
useemploy / utilize / leverage / adopt"utilize" is more formal than "use"; "leverage" emphasizes exploiting an advantage
aboutapproximately / roughly / circaUse "approximately" for numerical descriptions
tryattempt / endeavor
look atexamine / investigate / analyze / explore
find outdetermine / ascertain / identify / discover
go up / go downincrease / decrease / rise / decline
point outhighlight / emphasize / underscore
deal withaddress / tackle / handle / mitigate
make sureensure / verify / confirm
kind of / sort ofsomewhat / to some extent / partially
start / begininitiate / commence / undertake
end / finishconclude / terminate / complete
helpfacilitate / enable / assist / contribute to
needrequire / necessitate
canis capable of / is able to / has the potential toAvoid over-substitution — "can" is acceptable in academic writing

3.2 Vague → Precise Expression

Vague ExpressionPrecise AlternativeNotes
very good resultsstatistically significant improvement / a 12% increase in accuracyReplace vague modifiers with concrete data
some researchersSeveral studies (Chen et al., 2023; Li et al., 2024)Replace vague references with specific citations
recentlyIn the past five years / Since 2020Provide a time range
a fewthree / a small number of (n=3)Specify the quantity
it is known thatPrior work has established that (citation)Support with a citation
this is importantThis is critical for / This has significant implications forExplain why it matters

3.3 Reducing Redundancy

Redundant ExpressionConcise Version
in order toto
due to the fact thatbecause / since
at the present timecurrently / now
it is worth noting thatNotably, / Note that
it should be pointed out that(state the content directly)
a total of 50 samples50 samples
the vast majority ofmost
in the event thatif
has the ability tocan
on a daily basisdaily
in close proximity tonear
take into considerationconsider
is in agreement withagrees with
serves the function offunctions as

4. Voice Guidelines

4.1 Active vs. Passive Voice Selection

ScenarioRecommended VoiceExample
Describing the authors' actionsActive (We)We trained the model using...
Describing general methods/established factsPassiveThe data were collected from...
Emphasizing the object of an actionPassiveThe samples were analyzed using mass spectrometry.
Reporting resultsPrefer activeOur method achieves 95% accuracy.
Describing equipment/materialsPassiveThe solution was heated to 100°C.

4.2 Common Voice Issues

IssueIncorrect ExampleCorrection
Overuse of passiveIt was found by us that the results were improved by the method.We found that our method improved the results.
Inconsistent personThe author proposes... We then evaluate...Use "We" or "The authors" consistently
Meaningless passiveIt can be seen that accuracy increases.Accuracy increases. / The results show that accuracy increases.

4.3 Academic Person Conventions

PersonUse CaseNotes
WeDescribing the authors' own work (most common)Many journals accept "We" even for single-author papers
The authorsA more formal alternativeSome journals prefer this usage
ISingle-author theses and dissertationsSome journals do not accept this
OneGeneric/hypothetical statementsSomewhat old-fashioned; less common in modern academic writing

5. Coherence and Cohesion Rules

5.1 Intra-Paragraph Signaling Words

Logical RelationshipSignal Words/PhrasesExample
AdditionFurthermore, Moreover, Additionally, In additionFurthermore, our method generalizes well to unseen data.
ContrastHowever, In contrast, Conversely, On the other hand, NeverthelessHowever, this approach suffers from high computational cost.
Cause & EffectTherefore, Consequently, As a result, Hence, ThusTherefore, we adopt a two-stage training strategy.
ExemplificationFor example, For instance, Specifically, In particularSpecifically, we focus on the image classification task.
EmphasisIndeed, Notably, Importantly, It is worth noting thatNotably, the improvement is consistent across all datasets.
ConcessionAlthough, Despite, Notwithstanding, While, Even thoughAlthough the model is simple, it achieves competitive results.
SummaryIn summary, To summarize, Overall, In conclusionOverall, the proposed method outperforms existing baselines.
QualificationYet, Still, Nonetheless, That saidThat said, there are several limitations to our approach.
SequenceFirst, Second, Finally, Subsequently, ThenFirst, we preprocess the data. Subsequently, we train the model.
ConditionIf, Provided that, Given that, Assuming thatGiven that the dataset is imbalanced, we apply oversampling.

5.2 Inter-Paragraph Transition Patterns

PatternDescriptionExample Opening Sentence
HookThe end of one paragraph leads into the next topic"This raises the question of how to efficiently scale the model."
RecapThe next paragraph opens by revisiting the prior conclusion"Having established the effectiveness of our approach, we now turn to..."
Contrast BridgePoints out the shortcomings of the prior approach, introducing the current one"While these methods achieve reasonable accuracy, they fail to address..."
Question BridgeUses a question to create a transition"How can we overcome this limitation? In this section, we propose..."
Topic SentenceThe first sentence of each paragraph summarizes the core argument"The key advantage of our method is its ability to..."

5.3 Common Coherence Problems

ProblemDescriptionFix Strategy
Jumping argumentationLeaping from A to C without the B stepAdd intermediate reasoning steps or transitional sentences
Signal word overuseStarting every sentence with However / MoreoverReduce signal words; use sentence structure to convey logic
Signal word misuseUsing "Furthermore" to express contrastUse "However" for contrast; "Furthermore" for addition
Overly long paragraphsA single paragraph exceeding 8–10 sentencesSplit into 2–3 paragraphs by argument point
Overly short paragraphsA paragraph with only 1–2 sentencesMerge into a related paragraph or expand the discussion
Unclear reference"This shows..." — what does "this" refer to?"This result shows..." / "This finding indicates..."

6. Sentence Structure Optimization Rules

6.1 Strategies for Sentence Variety

StrategyOriginalImproved
Participial phrase openingWe use attention mechanism, and we improve accuracy.Leveraging the attention mechanism, we improve accuracy.
Inversion for emphasisThe improvement is particularly notable in low-resource settings.Particularly notable is the improvement in low-resource settings.
Appositive insertionThe model, which was proposed by Smith, achieves...The model, proposed by Smith (2023), achieves...
NominalizationWe improved the model, and this led to...The improvement of the model led to...
Parallel structureThe method is fast. It is also accurate. It is scalable too.The method is fast, accurate, and scalable.
Fronted adverbialAccuracy improved significantly when we added data augmentation.With data augmentation, accuracy improved significantly.

6.2 Sentence Structure Problems to Avoid

ProblemExampleFix
Overly long sentences (>40 words)We trained the model on the dataset which was collected from ... and preprocessed using ... and then evaluated on ...Split into 2–3 shorter sentences
Consecutive short sentencesThe accuracy is high. The model is fast. It uses less memory.Combine: The model achieves high accuracy with fast inference and low memory consumption.
Starting with There is/areThere are many studies that focus on...Many studies focus on...
Overuse of It is...that cleft sentencesIt is the attention mechanism that improves...The attention mechanism improves...
Noun pile-upsdeep learning image classification model performancethe performance of a deep learning model for image classification

7. Section-Specific Polishing Guide

7.1 Abstract

  • Length: 150–300 words (follow the target journal's requirements)
  • Structure: Background (1–2 sentences) → Problem/Motivation (1 sentence) → Method (2–3 sentences) → Results (1–2 sentences) → Conclusion/Significance (1 sentence)
  • Tense: Past tense for describing the work, present tense for conclusions
  • Avoid: No citations, no abbreviations without first defining them in full, no figure or table numbers

7.2 Introduction

  • Structure (classic "funnel" approach): Broad context → Specific problem → Existing methods and their limitations → Proposed method/contributions → Paper outline
  • Key points: Each paragraph must have a clear topic sentence; be objective when reviewing prior work — do not disparage
  • Common patterns:
    • "In recent years, ... has attracted increasing attention."
    • "Despite significant progress, ... remains a challenge."
    • "To address this issue, we propose..."
    • "The main contributions of this paper are as follows:"

7.3 Related Work

  • Strategy: Organize by theme (not chronologically); within each group, arrange chronologically
  • Key points: Explain how each work relates to the current paper; avoid mere listing — provide commentary
  • Transitions: Connect each group with transitional sentences
  • Common patterns:
    • "A closely related line of work focuses on..."
    • "In contrast to these approaches, our method..."
    • "Building upon the work of X, we extend..."

7.4 Methods

  • Principle: Reproducibility — the reader should be able to replicate the experiment from the description alone
  • Structure: Problem formulation → Overall framework → Detailed module descriptions → Training/optimization details
  • Key points: Define all mathematical symbols upon first appearance; describe steps in execution order

7.5 Results / Experiments

  • Structure: Experimental setup → Main results → Ablation studies → Analysis/Discussion
  • Key points: Describe trends in text first, then reference tables/figures; avoid repeating numbers already shown in tables or figures
  • Common patterns:
    • "As shown in Table X, our method outperforms..."
    • "We observe a consistent improvement of X% across..."
    • "The ablation study reveals that..."

7.6 Discussion

  • Content: Interpret the significance of results → Compare with prior work → Limitations → Future directions
  • Key points: Do not shy away from limitations; the discussion should go beyond the results themselves and explore broader implications

7.7 Conclusion

  • Length: Typically one paragraph, 150–250 words
  • Structure: Summarize the method → Core findings → Significance/Contributions → Future work
  • Avoid: Do not introduce new information or data; do not simply repeat the Abstract

8. Polishing Output Format

For each paragraph of text provided by the user, output in the following format:

### Original
[User's original text]

### Review

| Dimension | Rating | Key Issues |
|-----------|--------|-----------|
| Grammar | ✓ Good / △ Needs improvement / ✗ Significant issues | Brief description |
| Word Choice | ✓ / △ / ✗ | Brief description |
| Voice | ✓ / △ / ✗ | Brief description |
| Coherence | ✓ / △ / ✗ | Brief description |
| Sentence Structure | ✓ / △ / ✗ | Brief description |

### Detailed Changes
1. **Original**: "..."
   **Revised**: "..."
   **Reason**: [Specific rationale citing the rules above]

2. ...

### Polished Version
[Complete polished paragraph]

9. Domain-Specific Notes

Different disciplines have their own writing conventions. Respect field-specific norms when polishing:

FieldCharacteristicsNotes
Computer ScienceActive voice ("We") is commonMore colloquial phrasing is tolerated (e.g., "we run"); algorithm descriptions must be precise
Medicine/BiologyPassive voice predominates"Patients were randomized..."; terminology must conform to MeSH standards
PhysicsConcise, equation-drivenMathematical derivations must be rigorous; "one can show that..." is common
Social SciencesFrequent use of hedging"may," "might," "suggests"; avoid overly absolute statements
EngineeringResults-orientedEmphasis on performance metrics and experimental validation

10. Agent Behavior Guide

When the user submits text for polishing, follow this workflow:

  1. Confirm details: Paper type, target journal/conference (if any), polishing focus
  2. Identify the section: Determine which part of the paper the text belongs to and apply the corresponding section guidelines
  3. Five-dimension review: Check systematically: Grammar → Word Choice → Voice → Coherence → Sentence Structure
  4. Annotate sentence by sentence: Provide a reason for every change, citing the specific rule
  5. Output the polished version: Deliver the complete polished text
  6. Summarize recommendations: Highlight the main categories of issues and directions for improvement

Core Principles:

  • Preserve the author's original meaning and argumentation logic; do not alter technical content
  • Minimize changes — if a single word can be changed instead of a whole sentence, change only the word; if a sentence can be changed instead of a whole paragraph, change only the sentence
  • When uncertain whether something is an error, present it as a suggestion rather than making the change outright
  • Defer to the author's terminology unless it is clearly incorrect
  • Do not modify content the user has marked as "please keep"
Repository
zebbern/claude-code-guide
Last updated
First committed

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.