Purpose
This Skill provides a systematic, top-down workflow for polishing academic papers. It works from structure to logic to expression, with user confirmation at each decision point. Expression options are drawn from reference-driven academic patterns rather than ad hoc rewrites, ensuring professional and consistent output suitable for journal submission.
Trigger
Activates when the user asks to:
- Polish, revise, or improve an academic paper section by section
- 润色、精修、逐步改进学术论文
Example invocations:
- "Polish my paper section by section" / "润色论文"
- "Help me revise my introduction step by step" / "精修论文"
- "Guide me through polishing this draft" / "帮我逐步润色这篇论文"
Modes
| Mode |
Default |
Behavior |
interactive |
Yes |
Full 4-step flow with user confirmation at each decision point |
guided |
|
Multi-pass with confirmation at key checkpoints only |
direct |
|
Single-pass polish using defaults; skip AskUserQuestion |
batch |
|
Same operation applied across multiple sections sequentially |
Default mode: interactive
Mode inference: "quickly" or "just fix" switches to direct. "step by step" or "逐步" confirms interactive (already default).
References
Required (always loaded)
| File |
Purpose |
references/expression-patterns.md |
Academic expression patterns overview and module index |
Leaf Hints (loaded when needed)
| File |
When to Load |
references/expression-patterns/introduction-and-gap.md |
Polishing introduction or background content |
references/expression-patterns/methods-and-data.md |
Polishing methods, data, or study area content |
references/expression-patterns/results-and-discussion.md |
Polishing results or discussion content |
references/expression-patterns/conclusions-and-claims.md |
Polishing conclusion content |
references/expression-patterns/geography-domain.md |
Content involves spatial, urban, or planning topics |
references/anti-ai-patterns.md |
Polishing expression (Step 3) -- screen for AI-sounding phrases |
references/journals/ceus.md |
Target journal is CEUS |
Loading Rules
- Load expression patterns overview at start; select the appropriate leaf based on section type.
- Load anti-AI patterns when polishing expression (Step 3).
- Load journal template when target journal is specified.
- Load
geography-domain.md when spatial, urban, or planning content is detected.
- If a reference file is missing, warn the user and proceed with reduced capability.
Ask Strategy
Before starting, ask about:
- Target journal (if not already known)
- Which section to work on
- Preferred mode (if ambiguous from trigger)
Rules:
- Never ask more than 3 questions before producing initial output.
- In
direct mode, skip pre-questions if the user provided enough context.
- In
batch mode, skip per-item questions; apply settings from the first item.
- Use Structured Interaction when available; fall back to plain-text questions otherwise.
- See
skill-conventions.md > AskUserQuestion Enforcement for full rules.
Workflow
Step 1: Collect Context
- Determine input type (file path or pasted text).
- Load required references (expression-patterns overview).
- Identify target journal; load journal template if specified.
- Read input content using the Read tool; extract key numbers, claims, and data points.
- Locate example/reference papers if the user provides them (use Read tool for PDFs).
- In
interactive or guided mode: confirm scope with the user before proceeding.
Step 2: Structure & Logic Confirmation
- Analyze section macro structure (e.g., Abstract = Background + Gap + Method + Results + Contribution).
- Present a structure table for user confirmation.
- Break content into sentences; assign a logic function to each.
- Present the logic chain for user confirmation.
- Checkpoint: user confirms structure and logic before expression work begins.
- In
direct mode: run structure and logic analysis internally, proceed to Step 3 automatically.
Step 3: Expression Polish & Consistency
- Load the section-appropriate expression pattern leaf and anti-AI patterns.
- For each sentence with expression issues, present 2-3 options via AskUserQuestion (
interactive mode) or apply the best option automatically (direct mode).
AskUserQuestion({
question: "Which expression do you prefer for [sentence function]?",
options: [
{ label: "[Expression A]", description: "[full sentence with expression A]" },
{ label: "[Expression B]", description: "[full sentence with expression B]" },
{ label: "[Expression C]", description: "[full sentence with expression C]" }
]
})
- Reference paper consultation: when the user questions professionalism, use the Read tool to load example papers and extract expression patterns.
- Journal style check: apply journal-specific requirements from the loaded template.
- Repetition and coherence pass: check for repeated expressions and missing transitions; suggest fixes.
- Cross-section consistency: verify numbers, terminology, and claims across sections.
Step 4: Output
- Generate highlights if the journal requires them (for CEUS, see
references/journals/ceus.md).
- Suggest a read-aloud final check to catch awkward phrasing.
- Compile all confirmed content into the final version.
- Present the final version with word count for user confirmation.
- Write to
*_polished.md after confirmation (or automatically in batch mode).
- Report word count and any journal constraint notes.
Output Contract
| Output |
Format |
Condition |
polished_text |
Markdown file (*_polished.md) or conversation output |
Always produced |
change_summary |
Markdown in session |
Always produced |
| Word count |
Integer |
Always reported |
| Journal compliance notes |
Bullet list |
When a target journal is specified |
Edge Cases
| Situation |
Handling |
| Unprofessional word flagged |
Present 2-3 alternatives via AskUserQuestion; accept if user insists |
| Section too long for single pass |
Split into paragraph-level sub-passes; maintain cross-paragraph coherence |
| No journal specified |
Default to general academic style; note in output |
| Mixed language input |
Detect dominant language; ask user to confirm target language |
| Reference paper provided as PDF |
Use Read tool to load PDF; extract style patterns for expression matching |
| Abrupt sentence transition |
Provide transition options via AskUserQuestion |
| Repetition detected between sections |
Identify repeated content; suggest which occurrence to rephrase |
| Logic structure needs modification |
Return to Step 2 to re-confirm structure before continuing |
Fallbacks
| Scenario |
Fallback |
| Structured Interaction unavailable |
Ask 1-3 plain-text questions covering highest-impact gaps; do not block workflow |
| Reference file missing |
Log the missing file, proceed with reduced capability, warn the user |
| Target journal not specified |
Ask once; if declined, use general academic style |
| PDF reference paper unreadable |
Ask user to paste relevant excerpts instead |
Skill: paper-polish-workflow
Conventions: references/skill-conventions.md
1---2name: paper-polish-workflow3description: Systematic top-down workflow for polishing academic papers. Structure to logic to expression with user confirmation at each step.4---5
6## Purpose
7
8This Skill provides a systematic, top-down workflow for polishing academic papers. It works from structure to logic to expression, with user confirmation at each decision point. Expression options are drawn from reference-driven academic patterns rather than ad hoc rewrites, ensuring professional and consistent output suitable for journal submission.
9
10## Trigger
11
12**Activates when the user asks to:**
13- Polish, revise, or improve an academic paper section by section
14- 润色、精修、逐步改进学术论文
15
16**Example invocations:**
17- "Polish my paper section by section" / "润色论文"
18- "Help me revise my introduction step by step" / "精修论文"
19- "Guide me through polishing this draft" / "帮我逐步润色这篇论文"
20
21## Modes
22
23| Mode | Default | Behavior |
24|------|---------|----------|
25| `interactive` | Yes | Full 4-step flow with user confirmation at each decision point |
26| `guided` | | Multi-pass with confirmation at key checkpoints only |
27| `direct` | | Single-pass polish using defaults; skip AskUserQuestion |
28| `batch` | | Same operation applied across multiple sections sequentially |
29
30**Default mode:** `interactive`
31
32**Mode inference:** "quickly" or "just fix" switches to `direct`. "step by step" or "逐步" confirms `interactive` (already default).
33
34## References
35
36### Required (always loaded)
37
38| File | Purpose |
39|------|---------|
40| `references/expression-patterns.md` | Academic expression patterns overview and module index |
41
42### Leaf Hints (loaded when needed)
43
44| File | When to Load |
45|------|--------------|
46| `references/expression-patterns/introduction-and-gap.md` | Polishing introduction or background content |
47| `references/expression-patterns/methods-and-data.md` | Polishing methods, data, or study area content |
48| `references/expression-patterns/results-and-discussion.md` | Polishing results or discussion content |
49| `references/expression-patterns/conclusions-and-claims.md` | Polishing conclusion content |
50| `references/expression-patterns/geography-domain.md` | Content involves spatial, urban, or planning topics |
51| `references/anti-ai-patterns.md` | Polishing expression (Step 3) -- screen for AI-sounding phrases |
52| `references/journals/ceus.md` | Target journal is CEUS |
53
54### Loading Rules
55
56- Load expression patterns overview at start; select the appropriate leaf based on section type.
57- Load anti-AI patterns when polishing expression (Step 3).
58- Load journal template when target journal is specified.
59- Load `geography-domain.md` when spatial, urban, or planning content is detected.
60- If a reference file is missing, warn the user and proceed with reduced capability.
61
62## Ask Strategy
63
64**Before starting, ask about:**
651. Target journal (if not already known)
662. Which section to work on
673. Preferred mode (if ambiguous from trigger)
68
69**Rules:**
70- Never ask more than 3 questions before producing initial output.
71- In `direct` mode, skip pre-questions if the user provided enough context.
72- In `batch` mode, skip per-item questions; apply settings from the first item.
73- Use Structured Interaction when available; fall back to plain-text questions otherwise.
74- See `skill-conventions.md > AskUserQuestion Enforcement` for full rules.
75
76## Workflow
77
78### Step 1: Collect Context
79
80- Determine input type (file path or pasted text).
81- Load required references (expression-patterns overview).
82- Identify target journal; load journal template if specified.
83- Read input content using the Read tool; extract key numbers, claims, and data points.
84- Locate example/reference papers if the user provides them (use Read tool for PDFs).
85- In `interactive` or `guided` mode: confirm scope with the user before proceeding.
86
87### Step 2: Structure & Logic Confirmation
88
89- Analyze section macro structure (e.g., Abstract = Background + Gap + Method + Results + Contribution).
90- Present a structure table for user confirmation.
91- Break content into sentences; assign a logic function to each.
92- Present the logic chain for user confirmation.
93- Checkpoint: user confirms structure and logic before expression work begins.
94- In `direct` mode: run structure and logic analysis internally, proceed to Step 3 automatically.
95
96### Step 3: Expression Polish & Consistency
97
98- Load the section-appropriate expression pattern leaf and anti-AI patterns.
99- For each sentence with expression issues, present 2-3 options via AskUserQuestion (`interactive` mode) or apply the best option automatically (`direct` mode).
100
101```
102AskUserQuestion({
103 question: "Which expression do you prefer for [sentence function]?",
104 options: [
105 { label: "[Expression A]", description: "[full sentence with expression A]" },
106 { label: "[Expression B]", description: "[full sentence with expression B]" },
107 { label: "[Expression C]", description: "[full sentence with expression C]" }
108 ]
109})
110```
111
112- **Reference paper consultation:** when the user questions professionalism, use the Read tool to load example papers and extract expression patterns.
113- **Journal style check:** apply journal-specific requirements from the loaded template.
114- **Repetition and coherence pass:** check for repeated expressions and missing transitions; suggest fixes.
115- **Cross-section consistency:** verify numbers, terminology, and claims across sections.
116
117### Step 4: Output
118
119- Generate highlights if the journal requires them (for CEUS, see `references/journals/ceus.md`).
120- Suggest a read-aloud final check to catch awkward phrasing.
121- Compile all confirmed content into the final version.
122- Present the final version with word count for user confirmation.
123- Write to `*_polished.md` after confirmation (or automatically in `batch` mode).
124- Report word count and any journal constraint notes.
125
126## Output Contract
127
128| Output | Format | Condition |
129|--------|--------|-----------|
130| `polished_text` | Markdown file (`*_polished.md`) or conversation output | Always produced |
131| `change_summary` | Markdown in session | Always produced |
132| Word count | Integer | Always reported |
133| Journal compliance notes | Bullet list | When a target journal is specified |
134
135## Edge Cases
136
137| Situation | Handling |
138|-----------|----------|
139| Unprofessional word flagged | Present 2-3 alternatives via AskUserQuestion; accept if user insists |
140| Section too long for single pass | Split into paragraph-level sub-passes; maintain cross-paragraph coherence |
141| No journal specified | Default to general academic style; note in output |
142| Mixed language input | Detect dominant language; ask user to confirm target language |
143| Reference paper provided as PDF | Use Read tool to load PDF; extract style patterns for expression matching |
144| Abrupt sentence transition | Provide transition options via AskUserQuestion |
145| Repetition detected between sections | Identify repeated content; suggest which occurrence to rephrase |
146| Logic structure needs modification | Return to Step 2 to re-confirm structure before continuing |
147
148## Fallbacks
149
150| Scenario | Fallback |
151|----------|----------|
152| Structured Interaction unavailable | Ask 1-3 plain-text questions covering highest-impact gaps; do not block workflow |
153| Reference file missing | Log the missing file, proceed with reduced capability, warn the user |
154| Target journal not specified | Ask once; if declined, use general academic style |
155| PDF reference paper unreadable | Ask user to paste relevant excerpts instead |
156
157---
158
159*Skill: paper-polish-workflow*
160*Conventions: references/skill-conventions.md*