| name | method-polisher |
| description | Academic Method section polisher for research papers. Use this skill whenever the user wants to improve, restructure, evaluate, or rewrite the Method (or Methods, Methodology) section of a research paper. Triggers include: "polish my methods section", "is my methodology clear enough", "help me describe my procedure", "my methods section is too vague", or uploading a paper and asking to improve the methodology description. Also trigger when the user asks whether their method section has enough detail for replication. |
Method Polisher Skill
Polishes and restructures the Method section of research papers using Move-Step analysis. Analyzes the existing text, identifies gaps in coverage or clarity, and rewrites or refines accordingly. Default output language is English unless the user specifies otherwise.
Step 1: Read the Input
File input — use the first applicable method:
- Content already in context → process directly
- File reading tool available → use it
- Code execution available →
.docx: python-docx / .pdf: pdfplumber / .txt: open()
- Nothing available → ask user to paste the Method section text
After reading, identify:
- Research type: experimental (lab) vs. empirical (field/survey/observational) vs. computational
- Discipline (affects which Moves are expected and how detailed each should be)
- Current subheadings (Participants, Procedure, Materials, etc.)
Step 2: Diagnose the Existing Text
Map the current text against the four-Move framework. Identify which Moves and Steps are present, unclear, or missing.
Consult references/moves.md for detailed Move rules.
Diagnosis (share briefly with user before rewriting):
Move 1 (Describing subjects/sources): [present / vague / missing]
Move 2 (Describing experiment/procedure): [present / vague / missing]
Move 3 (Describing data): [present / vague / missing]
Move 4 (Introducing analysis method): [present / vague / missing]
Key issues: [list specific problems]
Step 3: Polish Using the Move Framework
The Method section must enable replication — every methodological decision that affects results must be stated. Read references/moves.md for full rules, then apply:
Move 1 — Describing subjects / data sources
- Who/what was studied (sample, corpus, dataset, materials)
- How they were selected or recruited
- Key characteristics (size, demographics, inclusion/exclusion criteria)
- Ethical considerations if applicable
Move 2 — Describing the experiment / procedure
- What was done, step by step, in chronological order
- Instruments, equipment, or stimuli used
- Controls and standardization measures
Move 3 — Describing the data
- What data was collected or recorded
- Data type, format, and size
- How data was recorded or coded
Move 4 — Introducing the method of analysis
- How data was analyzed (statistical tests, frameworks, software)
- Criteria or thresholds used
- Analytical framework or coding scheme
Polishing priorities:
- Check that Move 1 includes sampling rationale, not just sample description
- Ensure Move 2 is sequential and specific enough to be replicated
- Verify Move 3 defines key concepts/variables operationally
- Confirm Move 4 names specific analysis procedures (not just "data was analyzed")
- Apply consistent past tense and passive voice (see
references/language.md)
Step 4: Apply Language Standards
Consult references/language.md for tense, voice, and lexical signal rules.
Key rules:
- Dominant tense: past simple throughout (the method was carried out in the past)
- Dominant voice: passive (focus on what was done, not who did it)
- Active voice acceptable when the subject is clearly the research team
- Present simple only for definitions and established procedures that are still valid
Step 5: Output
Present the polished Method section with a brief note explaining:
- What structural gaps were filled
- What clarity or precision improvements were made
- Whether subheadings were added or reorganized
- Word count (typical: 300–800 words for a term paper; up to 1500 for a full article)
Self-check before output:
- Could another researcher replicate this study from this description?
- Is the sample/data source fully characterized (size, selection, characteristics)?
- Are all instruments, tools, and procedures named specifically?
- Are key concepts operationally defined?
- Is tense consistently past simple? Is passive voice used appropriately?
