manuscript-optimizer

Name: manuscript-optimizer
Author: Boom5426

62installs28forks

How to install manuscript-optimizer?

boom5426/nature-paper-skills/manuscript-optimizer

$npx -y skills add Boom5426/Nature-Paper-Skills --skill manuscript-optimizer

Installs into the current project.

Summary

Use when reviewing or revising an academic manuscript whose central claim, evidence chain, figures, terminology, and prose may have drifted out of sync before submission or resubmission.

Files · 1

View on GitHub

SKILL.md

1	# Manuscript Optimizer
2
3	## Overview
4
5	Use this skill to treat a manuscript like a precision instrument: fix the top-level design first, then the evidence chain, then the figures, then the terminology, and only then the sentence-level polish.
6
7	This workflow is not tied to a single paper or field. Use it across manuscript projects whenever structure, evidence, figures, and prose need to be brought back into alignment.
8
9	Core rule: do not spend effort polishing prose that sits on top of an unstable claim, a broken evidence chain, or inconsistent figures.
10
11	## When To Use
12
13	Use this skill when:
14	- A paper is being drafted, revised, resubmitted, or journal-adapted
15	- The abstract or introduction may be stronger than the downstream evidence
16	- The storyline feels diffuse, repetitive, or hard to defend
17	- Figures, legends, and main text may have drifted out of sync
18	- Core terminology or abbreviations may be unstable
19	- The writing needs to become clearer, tighter, and more reader-friendly without losing rigor
20
21	Do not use this skill as the primary workflow for:
22	- Pure literature review generation
23	- Citation-format-only cleanup
24	- Methods-only statistical review
25	- Journal peer review reports that focus mainly on acceptance recommendations
26
27	## Operating Principle
28
29	Always move in this order:
30	1. Direction first
31	2. Logic second
32	3. Visual evidence third
33	4. Terminology fourth
34	5. Language last
35
36	If a higher-level problem is unresolved, do not present lower-level polish as a solution.
37
38	## Two Modes
39
40	### Review Mode
41
42	Use when the task is to diagnose weaknesses before editing.
43
44	Output priorities:
45	- Findings first
46	- Highest-level issues first
47	- Explicitly separate unsupported claims, weak support, and cosmetic issues
48	- Cite exact sections, figures, or sentences when possible
49
50	### Optimization Mode
51
52	Use when the task is to actually rewrite or tighten the manuscript.
53
54	Execution order:
55	- Fix macro positioning and claim boundaries
56	- Repair section logic and evidence chain
57	- Sync figures, legends, and text
58	- Canonicalize terminology
59	- Polish prose, grammar, and format
60
61	## The Five-Level Audit
62
63	### 1. Top-Level Design And Core Contribution
64
65	Check the manuscript's top story before touching paragraph style.
66
67	Audit:
68	- What is the central problem?
69	- Why does it matter now?
70	- What is the single-sentence take-home message?
71	- Is the main contribution a method, framework, benchmark, resource, biological finding, or something else?
72	- Is the claim ambitious enough to matter but narrow enough to defend?
73	- After reading the abstract and introduction, can a broad scientific reader understand why this work is not interchangeable with prior work?
74
75	Guardrails:
76	- Do not let the paper sound like it contributes three equally important things unless that structure is deliberate and defensible
77	- Do not let examples, intuitions, or motivating cases masquerade as experimental evidence
78	- If the real contribution is a reformulation or evaluation framework, do not accidentally rewrite it as "a new model"
79
80	### 2. Logic Architecture And Evidence Chain
81
82	This is the main structural check.
83
84	Build a claim-to-evidence map:
85	- Extract every substantive claim from the abstract
86	- Extract every substantive claim from the introduction and discussion
87	- For each claim, point to the exact supporting result, figure, table, or supplementary item
88	- Mark each claim as:
89	- fully supported
90	- partially supported
91	- not supported by current evidence
92
93	Then run a reverse outline on the current section structure:
94	- Write the section thesis in one sentence.
95	- Write one line for each paragraph:
96	- paragraph job
97	- key evidence or reasoning inside it
98	- the transition relation to the previous paragraph
99	- Merge, move, or remove any paragraph that cannot be mapped cleanly to the section thesis.
100
101	When a claim is not fully supported, only three acceptable actions exist:
102	- weaken the claim
103	- add the missing evidence
104	- reframe the claim as intuition, hypothesis, or motivation
105
106	Questions to ask:
107	- Does each Results subsection answer a clear question?
108	- Does each module in the method or framework have a corresponding validation experiment?
109	- If the manuscript claims OOD generalization, cross-domain transfer, causal disentanglement, or clinical relevance, is there direct evidence for that exact statement?
110	- Are surprising or paradoxical findings explained, not merely reported?
111
112	Never leave the abstract or introduction stronger than the Results.
113
114	### Adversarial Self-Review
115
116	Before calling the structure stable, pressure-test the manuscript like a skeptical reviewer in five dimensions:
117	- contribution sufficiency
118	- writing clarity and reproducibility
119	- empirical strength
120	- evaluation completeness
121	- method or framework soundness
122
123	Do not answer these with intuition alone. P

People who liked this also viewed

6 picks

Type

manuscript-optimizer

62installs28forks

View on GitHub

How to install manuscript-optimizer?

boom5426/nature-paper-skills/manuscript-optimizer

$npx -y skills add Boom5426/Nature-Paper-Skills --skill manuscript-optimizer

Installs into the current project.

Summary

Use when reviewing or revising an academic manuscript whose central claim, evidence chain, figures, terminology, and prose may have drifted out of sync before submission or resubmission.

Files · 1

View on GitHub

SKILL.md

1	# Manuscript Optimizer
2
3	## Overview
4
5	Use this skill to treat a manuscript like a precision instrument: fix the top-level design first, then the evidence chain, then the figures, then the terminology, and only then the sentence-level polish.
6
7	This workflow is not tied to a single paper or field. Use it across manuscript projects whenever structure, evidence, figures, and prose need to be brought back into alignment.
8
9	Core rule: do not spend effort polishing prose that sits on top of an unstable claim, a broken evidence chain, or inconsistent figures.
10
11	## When To Use
12
13	Use this skill when:
14	- A paper is being drafted, revised, resubmitted, or journal-adapted
15	- The abstract or introduction may be stronger than the downstream evidence
16	- The storyline feels diffuse, repetitive, or hard to defend
17	- Figures, legends, and main text may have drifted out of sync
18	- Core terminology or abbreviations may be unstable
19	- The writing needs to become clearer, tighter, and more reader-friendly without losing rigor
20
21	Do not use this skill as the primary workflow for:
22	- Pure literature review generation
23	- Citation-format-only cleanup
24	- Methods-only statistical review
25	- Journal peer review reports that focus mainly on acceptance recommendations
26
27	## Operating Principle
28
29	Always move in this order:
30	1. Direction first
31	2. Logic second
32	3. Visual evidence third
33	4. Terminology fourth
34	5. Language last
35
36	If a higher-level problem is unresolved, do not present lower-level polish as a solution.
37
38	## Two Modes
39
40	### Review Mode
41
42	Use when the task is to diagnose weaknesses before editing.
43
44	Output priorities:
45	- Findings first
46	- Highest-level issues first
47	- Explicitly separate unsupported claims, weak support, and cosmetic issues
48	- Cite exact sections, figures, or sentences when possible
49
50	### Optimization Mode
51
52	Use when the task is to actually rewrite or tighten the manuscript.
53
54	Execution order:
55	- Fix macro positioning and claim boundaries
56	- Repair section logic and evidence chain
57	- Sync figures, legends, and text
58	- Canonicalize terminology
59	- Polish prose, grammar, and format
60
61	## The Five-Level Audit
62
63	### 1. Top-Level Design And Core Contribution
64
65	Check the manuscript's top story before touching paragraph style.
66
67	Audit:
68	- What is the central problem?
69	- Why does it matter now?
70	- What is the single-sentence take-home message?
71	- Is the main contribution a method, framework, benchmark, resource, biological finding, or something else?
72	- Is the claim ambitious enough to matter but narrow enough to defend?
73	- After reading the abstract and introduction, can a broad scientific reader understand why this work is not interchangeable with prior work?
74
75	Guardrails:
76	- Do not let the paper sound like it contributes three equally important things unless that structure is deliberate and defensible
77	- Do not let examples, intuitions, or motivating cases masquerade as experimental evidence
78	- If the real contribution is a reformulation or evaluation framework, do not accidentally rewrite it as "a new model"
79
80	### 2. Logic Architecture And Evidence Chain
81
82	This is the main structural check.
83
84	Build a claim-to-evidence map:
85	- Extract every substantive claim from the abstract
86	- Extract every substantive claim from the introduction and discussion
87	- For each claim, point to the exact supporting result, figure, table, or supplementary item
88	- Mark each claim as:
89	- fully supported
90	- partially supported
91	- not supported by current evidence
92
93	Then run a reverse outline on the current section structure:
94	- Write the section thesis in one sentence.
95	- Write one line for each paragraph:
96	- paragraph job
97	- key evidence or reasoning inside it
98	- the transition relation to the previous paragraph
99	- Merge, move, or remove any paragraph that cannot be mapped cleanly to the section thesis.
100
101	When a claim is not fully supported, only three acceptable actions exist:
102	- weaken the claim
103	- add the missing evidence
104	- reframe the claim as intuition, hypothesis, or motivation
105
106	Questions to ask:
107	- Does each Results subsection answer a clear question?
108	- Does each module in the method or framework have a corresponding validation experiment?
109	- If the manuscript claims OOD generalization, cross-domain transfer, causal disentanglement, or clinical relevance, is there direct evidence for that exact statement?
110	- Are surprising or paradoxical findings explained, not merely reported?
111
112	Never leave the abstract or introduction stronger than the Results.
113
114	### Adversarial Self-Review
115
116	Before calling the structure stable, pressure-test the manuscript like a skeptical reviewer in five dimensions:
117	- contribution sufficiency
118	- writing clarity and reproducibility
119	- empirical strength
120	- evaluation completeness
121	- method or framework soundness
122
123	Do not answer these with intuition alone. P

People who liked this also viewed

6 picks

Type