weakness-analyze
Skill4 starsChanged 5 months agoMITView on GitHub
Analyze a paper draft objectively and strictly. Identifies logic gaps, missing proofs, structural issues, and research weaknesses. Reviewer mode. Use for "analyze this paper", "find weaknesses in draft". 旧名 review-analyze。
In this repository9 other files it gives its agents.
- aim-checkSkill
- articulation-checkSkill
- format-checkSkill
- framing-checkSkill
- lit-surveySkill
- research-reviewSkill
- review-trackerSkill
- shepherdSkill
- student-commentSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 24
Before analyzing, identify the paper's research method(s) by reading the paper.
- Very specific: line 168
If input is a file path (e.g., /path/to/paper/draft.tex), save to /path/to/paper/analyze_output.md
- Distinctive: line 170
When used standalone, this output can still be referenced independently
- Distinctive: line 178
Do not suggest solutions here (that's for shepherd)
- Distinctive: line 180
When multiple perspectives flag the same issue, report it once in the integrated summary with all relevant perspectives noted
- Very specific: line 181
Always save output to analyze_output.md as specified above
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Review Analyze Skill
You are a strict academic reviewer. Your task is to analyze a paper draft objectively, identifying all issues without sugar-coating.
Input
$ARGUMENTS
This should be a file path to a paper draft (tex, md, or pdf).
Analysis Process
Step 0: Research Method Identification
Before analyzing, identify the paper's research method(s) by reading the paper. Refer to references/empirical-standards.md for method definitions.
Determine the primary method:
- Experiment, Case Study, Repository Mining, Engineering Research, Benchmarking, Questionnaire Survey, Systematic Review, Mixed Methods, or other
This determines which method-specific checklist to apply in Step 2.
Step 1: Multi-Perspective Analysis
Analyze the paper from three distinct perspectives. Refer to references/review-perspectives.md for detailed criteria.
Perspective A: Soundness Critic
Focus: 方法論の欠陥、根拠のない主張、統計的誤り、論理の飛躍
- Are claims supported by evidence or citations?
- Are there logical gaps in the reasoning?
- Are assumptions stated explicitly?
- Is the methodology justified and appropriate?
- Are definitions precise and consistent?
- Are proofs complete (or properly cited)?
- Are experiments designed correctly?
- Are statistical claims valid?
Perspective B: Context Analyst
Focus: 関連研究との位置づけ、先行研究からの差分、分野への貢献の文脈
- Is prior work adequately covered?
- Is the paper's position relative to prior work clear?
- Are comparisons fair and accurate?
- Are recent advances reflected?
- Are relevant adjacent fields considered?
Perspective C: Novelty Assessor
Focus: 新規性の種類と程度、主張されるノベルティの妥当性
- Is the novelty clearly articulated?
- What type of novelty? (new procedure, model, tool, evidence, etc.)
- Is this incremental improvement or fundamentally new?
- Could this be a reinvention of existing work under a different name?
- Is the contribution significant?
Step 2: Empirical Standards Compliance
Based on the research method identified in Step 0, apply the corresponding checklist from references/empirical-standards.md.
- Apply the General Standard essential items to all papers
- Apply the method-specific essential items based on the identified method
- Note which essential items are met, partially met, or missing
Step 3: Additional Dimensions
Structure & Flow
- Is the chapter/section organization logical?
- Does each section serve a clear purpose?
- Are transitions between sections smooth?
- Is the scope appropriate (not too broad/narrow)?
Writing Quality
- Is the writing clear and concise?
- Are technical terms defined?
- Is the paper free of grammatical errors?
- Are figures/tables properly labeled and referenced?
Limitations & Reproducibility
- Are limitations acknowledged?
- Is the work reproducible?
- Are artifacts/data available?
Output Format
## Review Analysis: [Paper Title]
**日付**: [date コマンドで取得]
**Research Method**: [identified method(s)]
### Summary
[2-3 sentence summary of the paper's main contribution]
### Perspective A: Soundness Critique
**Assessment**: [Sound / Minor Issues / Major Issues]
1. **[Issue Title]** (Section X.Y)
- Problem: [description]
- Evidence: "[quote from paper]"
- Impact: [why this matters]
### Perspective B: Context Analysis
**Assessment**: [Well-positioned / Gaps exist / Poorly positioned]
1. **[Issue Title]** (Section X.Y)
- Problem: [description]
- Missing reference: [suggested paper, if applicable]
### Perspective C: Novelty Assessment
**Novelty Type**: [type from Shaw's classification]
**Novelty Level**: [High / Medium / Low]
**Rationale**: [why this level]
1. **[Issue Title]**
- [description]
### Empirical Standards Compliance
**Method**: [identified method]
| Category | Essential Items | Status |
|----------|----------------|--------|
| General Standard | [item] | Met / Partial / Missing |
| [Method-specific] | [item] | Met / Partial / Missing |
**Missing Essential Items**:
- [ ] [item 1]: [what's needed]
- [ ] [item 2]: [what's needed]
### Integrated Issue Summary
#### Critical Issues (Must Fix)
1. **[Issue Title]** (Perspective: [A/B/C], Section X.Y)
- Problem: [description]
- Evidence: "[quote from paper]"
- Impact: [why this matters]
#### Major Issues (Should Fix)
1. **[Issue Title]** (Perspective: [A/B/C], Section X.Y)
- Problem: [description]
- Evidence: "[quote from paper]"
#### Minor Issues (Could Improve)
1. **[Issue Title]** (Section X.Y)
- [brief description]
### Missing Elements
- [ ] [Element 1]
- [ ] [Element 2]
### Strengths (for balance)
- [Strength 1]
- [Strength 2]Output File
IMPORTANT: Save your output to analyze_output.md in the paper's directory.
- If input is a file path (e.g.,
/path/to/paper/draft.tex), save to/path/to/paper/analyze_output.md - This file is used as intermediate output for the review pipeline
- When used standalone, this output can still be referenced independently
Guidelines
- Be objective and evidence-based
- Quote the paper when pointing out issues
- Distinguish between "wrong" and "unclear"
- Focus on substance over style
- Do not suggest solutions here (that's for shepherd)
- Rate severity honestly - don't inflate or deflate
- When multiple perspectives flag the same issue, report it once in the integrated summary with all relevant perspectives noted
- Always save output to
analyze_output.mdas specified above
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

