brycewang-stanford/Auto-Empirical-Research-Skills / skills/01-lishix520-academic-paper-skills/composer
academic-paper-composer
UnverifiedSystematic writing framework for philosophy and interdisciplinary academic papers from optimized outline to submission-ready manuscript. Use when users want to: (1) write a paper from a detailed outline, (2) ensure quality control during writing, (3) maintain consistency across chapters, (4) prepare a submission-ready manuscript, or (5) systematically execute a planned paper. Triggered by phrases like 'write the paper from this outline,' 'compose the full manuscript,' 'execute the outline,' or when users have completed strategic planning (academic-paper-strategist skill) and are ready to write. Takes optimized outline as input; outputs complete manuscript with iterative quality checks.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 25.7 KB |
| references/section_guides.md | markdown | 21.5 KB |
| references/writing_standards.md | markdown | 17.0 KB |
references/writing_standards.md
View on GitHub<!--
╔══════════════════════════════════════════════════════════════╗
║ 本文件为开源 Skill 原始文档,收录仅供学习与研究参考 ║
║ CoPaper.AI 收集整理 | https://copaper.ai ║
╚══════════════════════════════════════════════════════════════╝
来源仓库: https://github.com/lishix520/academic-paper-skills
项目名称: academic-paper-skills
开源协议: MIT License
收录日期: 2026-04-02
声明: 本文件版权归原作者所有。此处收录旨在为社会科学实证研究者
提供 AI Agent Skills 的集中参考。如有侵权,请联系删除。
-->
# Writing Quality Standards for Academic Papers
This document defines quality evaluation criteria for systematic academic paper writing.
## Overview
Quality evaluation occurs at two levels:
1. **Chapter-level quality checks** (after each chapter is written)
2. **Final quality assessment** (before submission)
---
## Chapter-Level Quality Standards
### 5-Dimension Assessment (4 points each, 20 total)
After writing each chapter, evaluate against these 5 dimensions:
#### 1. Argument Quality (1-4 points)
**4 points - Excellent**:
- Thesis/claim crystal clear
- All supporting arguments logically connected
- Premises fully justified
- No logical gaps or unsupported assertions
- Counter-arguments addressed where appropriate
**3 points - Good**:
- Clear main argument
- Most support adequate
- Minor gaps that could be strengthened
- Some premises could use more justification
**2 points - Acceptable**:
- Argument identifiable but could be clearer
- Multiple gaps in reasoning
- Several premises need justification
- Logical flow interrupted
**1 point - Needs Major Revision**:
- Unclear argument
- Major logical gaps
- Many unsupported claims
- Disjointed reasoning
#### 2. Citation Quality (1-4 points)
**4 points - Excellent**:
- All claims properly cited
- Citations accurate and relevant
- Key literature well-integrated
- Citation density appropriate for section type
- Format consistent throughout
**3 points - Good**:
- Most claims cited
- Citations generally relevant
- Minor formatting inconsistencies
- Could integrate 1-2 more key sources
**2 points - Acceptable**:
- Some claims lack citations
- Several citations not optimally relevant
- Format inconsistencies present
- Missing some key literature
**1 point - Needs Major Revision**:
- Many uncited claims
- Citations poorly integrated
- Major format issues
- Key literature missing
#### 3. Clarity & Readability (1-4 points)
**4 points - Excellent**:
- Prose clear and precise
- Technical terms defined when introduced
- Sentence structure varied and effective
- Transitions smooth between ideas
- No ambiguity in meaning
**3 points - Good**:
- Generally clear prose
- Most terms adequately explained
- Occasional awkward phrasing
- Mostly good transitions
**2 points - Acceptable**:
- Some unclear passages
- Several terms need better explanation
- Multiple awkward sentences
- Transitions need improvement
**1 point - Needs Major Revision**:
- Frequently unclear
- Many unexplained terms
- Poor sentence structure
- Weak or missing transitions
#### 4. Structure & Flow (1-4 points)
**4 points - Excellent**:
- Perfect logical progression
- Each paragraph serves clear purpose
- Subsections well-organized
- Chapter fits outline specification exactly
- Word count within ±5% of target
**3 points - Good**:
- Good overall flow
- Most paragraphs purposeful
- Organization generally clear
- Minor deviations from outline
- Word count within ±10%
**2 points - Acceptable**:
- Flow interrupted in places
- Some paragraphs unfocused
- Organization could be improved
- Notable deviations from outline
- Word count off by 11-20%
**1 point - Needs Major Revision**:
- Poor flow
- Many unfocused paragraphs
- Disorganized
- Major deviations from outline
- Word count off by >20%
#### 5. Platform Style Conformity (1-4 points)
**4 points - Excellent**:
- Perfect match to platform writing style
- Voice consistent with sample papers
- Terminology appropriate
- Format conventions followed exactly
- Citation style matches platform
**3 points - Good**:
- Good style match
- Minor voice inconsistencies
- Terminology mostly appropriate
- Format mostly correct
**2 points - Acceptable**:
- Recognizable platform style but inconsistent
- Several voice/terminology issues
- Format needs adjustment in places
**1 point - Needs Major Revision**:
- Poor style match
- Wrong voice or terminology
- Format significantly different from platform norms
### Chapter Quality Gate
**Passing Threshold**: ≥16/20 (80%)
**If chapter scores <16/20**:
- Identify specific dimension(s) scoring <3
- Implement targeted revisions
- Re-evaluate before proceeding to next chapter
**Why This Matters**: Each chapter builds on previous ones. Low-quality chapters create cascading problems that are harder to fix later.
---
## Section-Specific Quality Standards
### Abstract Quality (250-300 words)
**Required Elements**:
- ✓ Problem/puzzle statement (1-2 sentences)
- ✓ Existing approach limitations (1-2 sentences)
- ✓ This paper's contribution (2-3 sentences)
- ✓ Main argument/findings (2-3 sentences)
- ✓ Significance/implications (1-2 sentences)
**Quality Checklist**:
- [ ] Can be understood without reading paper
- [ ] No citations in abstract (platform-dependent)
- [ ] Avoids jargon or defines essential terms
- [ ] Accurately represents paper content
- [ ] Compelling to target audience
### Introduction Quality (typically 15-20% of paper)
**Required Elements**:
- ✓ Opening puzzle/problem (hooks reader)
- ✓ Background context (enough to understand problem)
- ✓ Literature review (what's been tried, what's missing)
- ✓ This paper's approach (clear statement)
- ✓ Contribution summary (what's new)
- ✓ Roadmap (chapter-by-chapter preview)
**Quality Checklist**:
- [ ] Problem significance clearly established
- [ ] Gap in literature explicitly identified
- [ ] Originality clearly articulated
- [ ] Roadmap matches actual chapter structure
- [ ] Appropriate citation density (15-25 citations typical for philosophy)
### Main Body Chapters (60-70% of paper)
**Quality Varies by Chapter Type**:
**Conceptual/Theoretical Chapters**:
- [ ] Key concepts clearly defined
- [ ] Distinctions drawn where needed
- [ ] Examples provided for abstract concepts
- [ ] Connections to literature explicit
**Argument Chapters**:
- [ ] Premises clearly stated
- [ ] Inference steps explicit
- [ ] Objections anticipated and addressed
- [ ] Conclusion follows logically
**Case Study/Analysis Chapters**:
- [ ] Case described in sufficient detail
- [ ] Analysis systematically applied
- [ ] Findings clearly presented
- [ ] Implications drawn
**Critical Review Chapters**:
- [ ] Target position accurately represented
- [ ] Criticism fair and well-supported
- [ ] Charitable interpretation maintained
- [ ] Alternative considered
### Conclusion Quality (10-15% of paper)
**Required Elements**:
- ✓ Thesis restatement (in light of arguments)
- ✓ Main findings summary
- ✓ Significance reiteration (why this matters)
- ✓ Limitations acknowledgment (honest assessment)
- ✓ Future work directions (2-3 concrete suggestions)
**Quality Checklist**:
- [ ] No new information introduced
- [ ] Addresses introduction promises
- [ ] Acknowledges limitations without undermining contribution
- [ ] Future work suggestions are specific and feasible
- [ ] Strong closing statement
---
## Final Quality Assessment
### 7-Dimension Evaluation (10 points each, 70 total)
Before submission, evaluate complete paper on 7 dimensions:
#### 1. Overall Argument Quality (1-10)
**9-10 - Excellent**:
- Thesis perfectly clear throughout
- All chapters contribute to central argument
- Premises fully justified
- No logical gaps
- Counter-arguments thoroughly addressed
- Conclusion strongly supported
**7-8 - Good**:
- Clear thesis
- Good chapter integration
- Minor logical gaps
- Most objections addressed
**5-6 - Acceptable**:
- Thesis identifiable but could be clearer
- Some chapter disconnects
- Several logical gaps
- Some objections unaddressed
**3-4 - Needs Revision**:
- Unclear thesis
- Poor chapter integration
- Major logical gaps
**1-2 - Major Revision Required**:
- No clear thesis
- Chapters poorly connected
- Pervasive logical problems
#### 2. Literature Integration (1-10)
**9-10 - Excellent**:
- 40-60 sources (typical for philosophy)
- All key literature covered
- Citations well-integrated
- Literature critically engaged
- Gap clearly positioned
**7-8 - Good**:
- 30-40 sources
- Most key works included
- Good integration
- Critical engagement present
**5-6 - Acceptable**:
- 20-30 sources
- Some key works missing
- Adequate integration
- Limited critical engagement
**3-4 - Needs Revision**:
- <20 sources
- Major gaps in literature
- Poor integration
**1-2 - Major Revision Required**:
- Insufficient sources
- Key literature ignored
- Citations merely decorative
#### 3. Clarity & Accessibility (1-10)
**9-10 - Excellent**:
- Consistently clear prose
- Complex ideas well-explained
- Appropriate for target audience
- No ambiguity
- Technical terms defined
**7-8 - Good**:
- Generally clear
- Most ideas accessible
- Minor unclear passages
**5-6 - Acceptable**:
- Clarity inconsistent
- Several unclear sections
- Some inaccessible passages
**3-4 - Needs Revision**:
- Frequently unclear
- Many inaccessible passages
**1-2 - Major Revision Required**:
- Pervasively unclear
- Inaccessible to target audience
#### 4. Originality & Contribution (1-10)
**9-10 - Excellent**:
- Major original contribution
- Clear advance over literature
- Significance well-established
- Multiple innovative elements
**7-8 - Good**:
- Clear original contribution
- Solid advance over literature
- Significance demonstrated
**5-6 - Acceptable**:
- Identifiable contribution
- Incremental advance
- Significance could be stronger
**3-4 - Needs Revision**:
- Unclear contribution
- Minimal advance
**1-2 - Major Revision Required**:
- No clear contribution
- Replicates existing work
#### 5. Methodological Rigor (1-10)
**9-10 - Excellent**:
- Method explicitly stated and justified
- Consistently applied
- Appropriate for research question
- Limitations acknowledged
- Alternative methods considered
**7-8 - Good**:
- Method clear and appropriate
- Generally consistent application
- Limitations noted
**5-6 - Acceptable**:
- Method identifiable
- Some inconsistency
- Limitations minimally addressed
**3-4 - Needs Revision**:
- Method unclear
- Inconsistent application
**1-2 - Major Revision Required**:
- No clear method
- Inappropriate approach
#### 6. Structure & Organization (1-10)
**9-10 - Excellent**:
- Perfect logical flow
- Optimal chapter organization
- Proportions balanced
- Transitions seamless
- Roadmap followed exactly
**7-8 - Good**:
- Good flow
- Sensible organization
- Proportions generally balanced
- Good transitions
**5-6 - Acceptable**:
- Flow adequate
- Organization acceptable
- Some proportion issues
- Transitions need work
**3-4 - Needs Revision**:
- Poor flow
- Questionable organization
- Imbalanced proportions
**1-2 - Major Revision Required**:
- No clear flow
- Disorganized
- Severely imbalanced
#### 7. Platform & Style Conformity (1-10)
**9-10 - Excellent**:
- Perfect platform style match
- Format exactly correct
- Voice consistent throughout
- Citation format perfect
- Length optimal
**7-8 - Good**:
- Good style match
- Format mostly correct
- Voice generally consistent
- Minor format issues
**5-6 - Acceptable**:
- Recognizable platform style
- Format needs adjustment
- Voice inconsistencies
**3-4 - Needs Revision**:
- Poor style match
- Format problems
**1-2 - Major Revision Required**:
- Wrong platform style
- Format significantly incorrect
### Final Quality Gate
**Passing Threshold**: ≥56/70 (80%)
**Score Interpretation**:
- 63-70 (90-100%): Excellent, ready for submission
- 56-62 (80-89%): Good, minor revisions recommended
- 49-55 (70-79%): Acceptable, moderate revisions needed
- <49 (<70%): Major revisions required
---
## Content Completeness Checklist
Before final submission, verify all required elements present:
### Structural Completeness
- [ ] Abstract (250-300 words)
- [ ] Introduction with all required elements
- [ ] All outlined main chapters present
- [ ] Conclusion with all required elements
- [ ] References section formatted correctly
### Content Completeness
- [ ] All promises from introduction fulfilled
- [ ] All claims supported by evidence or argument
- [ ] All technical terms defined
- [ ] All objections addressed
- [ ] All limitations acknowledged
### Citation Completeness
- [ ] Every citation in text has bibliography entry
- [ ] Every bibliography entry is cited in text
- [ ] Citation format consistent throughout
- [ ] All citations include necessary information (author, year, page if direct quote)
### Format Completeness
- [ ] Title page (if required by platform)
- [ ] Section numbering consistent
- [ ] Heading hierarchy logical
- [ ] Figure/table captions (if applicable)
- [ ] Appendices (if needed)
### Submission Checklist (Platform-Specific)
**PhilArchive/PhilPapers**:
- [ ] PDF format
- [ ] Abstract <500 words
- [ ] Proper metadata (title, keywords, AMS classification)
- [ ] Author information complete
**arXiv**:
- [ ] LaTeX or PDF format
- [ ] Abstract <1920 characters
- [ ] Proper category selection (cs.AI, q-bio.NC, etc.)
- [ ] No embedded fonts issues
**PhilSci-Archive**:
- [ ] PDF format
- [ ] Proper subject classification
- [ ] Keywords provided
- [ ] No copywritten material without permission
---
## Common Quality Issues and Fixes
### Issue 1: Argument Gaps
**Symptom**: Chapter scores low on "Argument Quality"
**Diagnosis**: Logical leaps, unsupported premises, or missing steps
**Fix**:
1. Identify specific gap (which premise or inference?)
2. Add explicit justification for that step
3. Consider if additional sub-argument needed
4. Re-evaluate argument flow
### Issue 2: Insufficient Citations
**Symptom**: Chapter scores low on "Citation Quality"
**Diagnosis**: Claims without support, missing key literature
**Fix**:
1. Review each paragraph - which claims need citations?
2. Search for supporting literature for uncited claims
3. Check if key works from literature review are actually cited
4. Aim for 10-15 citations per 1000 words (adjust by field)
### Issue 3: Unclear Prose
**Symptom**: Chapter scores low on "Clarity & Readability"
**Diagnosis**: Complex sentences, undefined terms, weak transitions
**Fix**:
1. Break complex sentences into simpler ones
2. Add definitions for technical terms at first use
3. Add transition sentences between paragraphs
4. Read aloud to identify awkward phrasing
### Issue 4: Structural Problems
**Symptom**: Chapter scores low on "Structure & Flow"
**Diagnosis**: Poor organization, wrong proportions, deviates from outline
**Fix**:
1. Create reverse outline (what does each paragraph actually do?)
2. Reorganize paragraphs for better flow
3. Cut or expand sections to match target proportions
4. Ensure chapter follows outline specification
### Issue 5: Style Mismatch
**Symptom**: Chapter scores low on "Platform Style Conformity"
**Diagnosis**: Wrong voice, inappropriate terminology, format issues
**Fix**:
1. Review platform writing standards guide
2. Check sample papers for voice and terminology
3. Adjust sentence structure to match platform norms
4. Verify citation format matches platform requirements
---
## Quality Improvement Strategies
### Strategy 1: Iterative Refinement
For chapters scoring 14-15/20:
- Identify the single lowest-scoring dimension
- Focus revision effort on that dimension only
- Re-evaluate
- If still <16/20, address next-lowest dimension
### Strategy 2: Peer Review Simulation
After writing each chapter:
1. Read as if you're a skeptical reviewer
2. List every question or objection that arises
3. Address each in revision
4. Re-evaluate
### Strategy 3: Cross-Chapter Coherence
After completing all main chapters:
1. Create concept map across chapters
2. Verify terminology consistent throughout
3. Check that chapter N builds on chapter N-1
4. Ensure conclusion addresses introduction promises
### Strategy 4: Platform Benchmarking
Before final evaluation:
1. Re-read 2-3 sample papers from target platform
2. Note their structural and stylistic patterns
3. Compare your paper to these patterns
4. Adjust to match where needed
---
## Summary
**Chapter-Level Quality**:
- 5 dimensions, 4 points each (20 total)
- Passing threshold: ≥16/20 (80%)
- Evaluate after each chapter before proceeding
**Final Quality**:
- 7 dimensions, 10 points each (70 total)
- Passing threshold: ≥56/70 (80%)
- Evaluate complete paper before submission
**Iterative Improvement**:
- Low-scoring chapters must be revised before proceeding
- Final paper must meet threshold before submission
- Use targeted revision strategies for specific issues
This ensures systematic quality control throughout the writing process.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |