ppw-logic

Name: ppw-logic
Author: Lylll9436

379stars20forks

How to install ppw-logic?

lylll9436/paper-polish-workflow-skill/ppw-logic

$npx -y skills add Lylll9436/Paper-Polish-Workflow-skill --skill ppw-logic

Installs into the current project.

Summary

Verify logical consistency across paper sections. Traces argument chains and identifies gaps, unsupported claims, terminology inconsistencies, and number contradictions. 论文逻辑验证，识别论证链断裂、无支撑声明、术语不一致、数字矛盾。

Files · 1

View on GitHub

SKILL.md

1	## Purpose
2
3	This Skill reads a full academic paper and verifies logical consistency across its sections. It identifies four types of issues — argument chain gaps (AC-), unsupported claims (UC-), terminology inconsistencies (TI-), and number contradictions (NC-) — and produces a two-part bilingual report. Part 1 maps each section's primary claim as an Argument Chain View table, showing how claims connect (or fail to connect) from Introduction through Conclusion. Part 2 categorizes every identified issue with a problem description, an impact statement, and a one-sentence directional suggestion, each followed by an inline Chinese translation. The Skill never rewrites text; it identifies and suggests only.
4
5	## Core Prompt
6
7	> Source: [awesome-ai-research-writing](https://github.com/Leey21/awesome-ai-research-writing) — 逻辑检查
8
9	````markdown
10	# Role
11	你是一位负责论文终稿校对的学术助手。你的任务是进行"红线审查"，确保论文没有致命错误。
12
13	# Task
14	请对我提供的【英文 LaTeX 代码片段】进行最后的一致性与逻辑核对。
15
16	# Constraints
17	1. 审查阈值（高容忍度）：
18	- 默认假设：请预设当前的草稿已经经过了多轮修改与校正，质量较高。
19	- 仅报错原则：只有在遇到阻碍读者理解的逻辑断层、引起歧义的术语混乱、或严重的语法错误时才提出意见。
20	- 严禁优化：对于"可改可不改"的风格问题、或者仅仅是"换个词听起来更高级"的建议，请直接忽略，不要通过挑刺来体现你的存在感。
21
22	2. 审查维度：
23	- 致命逻辑：是否存在前后完全矛盾的陈述？
24	- 术语一致性：核心概念是否在没有说明的情况下换了名字？
25	- 严重语病：是否存在导致句意不清的中式英语（Chinglish）或语法结构错误。
26
27	3. 输出格式：
28	- 如果没有上述"必须修改"的错误，请直接输出中文：[检测通过，无实质性问题]。
29	- 如果有问题，请使用中文分点简要指出，不要长篇大论。
30	````
31
32	## Trigger
33
34	Activates when the user asks to:
35	- Check, verify, or find logical problems in their paper
36	- 检查论文逻辑、验证论证链、帮我找逻辑问题
37
38	Example invocations:
39	- "Check the logic of my paper" / "检查我的论文逻辑"
40	- "Verify argument chain across sections" / "验证论文各章节论证是否自洽"
41	- "Find logical inconsistencies in my draft" / "帮我找论文里的逻辑问题"
42
43	## Modes
44
45	\| Mode \| Default \| Behavior \|
46	\|------\|---------\|----------\|
47	\| `direct` \| Yes \| Single-pass full-paper analysis, no intermediate confirmation steps \|
48	\| `batch` \| \| Not supported — cross-section logic verification requires full-paper context \|
49
50	Default mode: `direct`. User provides the paper and receives a complete two-part logic report.
51
52	## References
53
54	### Required (always loaded)
55
56	Logic verification loads `references/bilingual-output.md` to govern the Chinese translation format and opt-out behavior. No expression pattern leaves are needed, and anti-AI patterns are not loaded (consistent with Reviewer Simulation Skill convention). Logic verification is also journal-agnostic; no journal templates are loaded.
57
58	### Leaf Hints
59
60	None.
61
62	## Ask Strategy
63
64	Before starting, ask about:
65	1. Input format: file path or pasted text? (skip if obvious from how the user invoked the Skill)
66	2. Section structure: ask only if section boundaries are unclear from the input (e.g., unnumbered prose without headings)
67
68	Rules:
69	- In `direct` mode with sufficient input, proceed without pre-questions.
70	- Do not ask more than 1 question before starting analysis.
71	- Use Structured Interaction when available; fall back to a single plain-text question otherwise.
72
73	## Workflow
74
75	### Step 0: Workflow Memory Check
76
77	- Read `.planning/workflow-memory.json`. If file missing or empty, skip to Step 1.
78	- Check if the last 1-2 log entries form a recognized pattern with `ppw:logic` that has appeared >= threshold times in the log. See `skill-conventions.md > Workflow Memory > Pattern Detection` for the full algorithm.
79	- If a pattern is found, present recommendation via AskUserQuestion:
80	- Question: "检测到常用流程：[pattern]（已出现 N 次）。是否直接以 direct 模式运行 ppw:logic？"
81	- Options: "Yes, proceed" / "No, continue normally"
82	- If user accepts: set mode to `direct`, skip Ask Strategy questions.
83	- If user declines or AskUserQuestion unavailable: continue in normal mode.
84
85	### Step 1 — Input Guard
86
87	- Accept input as a file path (Read tool) or pasted text from conversation.
88	- Full-paper guard: If input appears to be partial — single section heading only, fewer than ~500 words, or missing Methods/Results/Discussion markers — refuse with:
89	> "Cross-section verification requires the full paper. Please provide the complete manuscript."
90	- Parse section structure: identify section boundaries and titles exactly as they appear in the paper.
91	- Record workflow: Append `{"skill": "ppw:logic", "ts": "<ISO timestamp>"}` to `.planning/workflow-memory.json`. Create file as `[]` if missing. Drop oldest entry if log length >= 50.

People who liked this also viewed

6 picks

Type

ppw-logic

379stars20forks

View on GitHub

How to install ppw-logic?

lylll9436/paper-polish-workflow-skill/ppw-logic

$npx -y skills add Lylll9436/Paper-Polish-Workflow-skill --skill ppw-logic

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	## Purpose
2
3	This Skill reads a full academic paper and verifies logical consistency across its sections. It identifies four types of issues — argument chain gaps (AC-), unsupported claims (UC-), terminology inconsistencies (TI-), and number contradictions (NC-) — and produces a two-part bilingual report. Part 1 maps each section's primary claim as an Argument Chain View table, showing how claims connect (or fail to connect) from Introduction through Conclusion. Part 2 categorizes every identified issue with a problem description, an impact statement, and a one-sentence directional suggestion, each followed by an inline Chinese translation. The Skill never rewrites text; it identifies and suggests only.
4
5	## Core Prompt
6
7	> Source: [awesome-ai-research-writing](https://github.com/Leey21/awesome-ai-research-writing) — 逻辑检查
8
9	````markdown
10	# Role
11	你是一位负责论文终稿校对的学术助手。你的任务是进行"红线审查"，确保论文没有致命错误。
12
13	# Task
14	请对我提供的【英文 LaTeX 代码片段】进行最后的一致性与逻辑核对。
15
16	# Constraints
17	1. 审查阈值（高容忍度）：
18	- 默认假设：请预设当前的草稿已经经过了多轮修改与校正，质量较高。
19	- 仅报错原则：只有在遇到阻碍读者理解的逻辑断层、引起歧义的术语混乱、或严重的语法错误时才提出意见。
20	- 严禁优化：对于"可改可不改"的风格问题、或者仅仅是"换个词听起来更高级"的建议，请直接忽略，不要通过挑刺来体现你的存在感。
21
22	2. 审查维度：
23	- 致命逻辑：是否存在前后完全矛盾的陈述？
24	- 术语一致性：核心概念是否在没有说明的情况下换了名字？
25	- 严重语病：是否存在导致句意不清的中式英语（Chinglish）或语法结构错误。
26
27	3. 输出格式：
28	- 如果没有上述"必须修改"的错误，请直接输出中文：[检测通过，无实质性问题]。
29	- 如果有问题，请使用中文分点简要指出，不要长篇大论。
30	````
31
32	## Trigger
33
34	Activates when the user asks to:
35	- Check, verify, or find logical problems in their paper
36	- 检查论文逻辑、验证论证链、帮我找逻辑问题
37
38	Example invocations:
39	- "Check the logic of my paper" / "检查我的论文逻辑"
40	- "Verify argument chain across sections" / "验证论文各章节论证是否自洽"
41	- "Find logical inconsistencies in my draft" / "帮我找论文里的逻辑问题"
42
43	## Modes
44
45	\| Mode \| Default \| Behavior \|
46	\|------\|---------\|----------\|
47	\| `direct` \| Yes \| Single-pass full-paper analysis, no intermediate confirmation steps \|
48	\| `batch` \| \| Not supported — cross-section logic verification requires full-paper context \|
49
50	Default mode: `direct`. User provides the paper and receives a complete two-part logic report.
51
52	## References
53
54	### Required (always loaded)
55
56	Logic verification loads `references/bilingual-output.md` to govern the Chinese translation format and opt-out behavior. No expression pattern leaves are needed, and anti-AI patterns are not loaded (consistent with Reviewer Simulation Skill convention). Logic verification is also journal-agnostic; no journal templates are loaded.
57
58	### Leaf Hints
59
60	None.
61
62	## Ask Strategy
63
64	Before starting, ask about:
65	1. Input format: file path or pasted text? (skip if obvious from how the user invoked the Skill)
66	2. Section structure: ask only if section boundaries are unclear from the input (e.g., unnumbered prose without headings)
67
68	Rules:
69	- In `direct` mode with sufficient input, proceed without pre-questions.
70	- Do not ask more than 1 question before starting analysis.
71	- Use Structured Interaction when available; fall back to a single plain-text question otherwise.
72
73	## Workflow
74
75	### Step 0: Workflow Memory Check
76
77	- Read `.planning/workflow-memory.json`. If file missing or empty, skip to Step 1.
78	- Check if the last 1-2 log entries form a recognized pattern with `ppw:logic` that has appeared >= threshold times in the log. See `skill-conventions.md > Workflow Memory > Pattern Detection` for the full algorithm.
79	- If a pattern is found, present recommendation via AskUserQuestion:
80	- Question: "检测到常用流程：[pattern]（已出现 N 次）。是否直接以 direct 模式运行 ppw:logic？"
81	- Options: "Yes, proceed" / "No, continue normally"
82	- If user accepts: set mode to `direct`, skip Ask Strategy questions.
83	- If user declines or AskUserQuestion unavailable: continue in normal mode.
84
85	### Step 1 — Input Guard
86
87	- Accept input as a file path (Read tool) or pasted text from conversation.
88	- Full-paper guard: If input appears to be partial — single section heading only, fewer than ~500 words, or missing Methods/Results/Discussion markers — refuse with:
89	> "Cross-section verification requires the full paper. Please provide the complete manuscript."
90	- Parse section structure: identify section boundaries and titles exactly as they appear in the paper.
91	- Record workflow: Append `{"skill": "ppw:logic", "ts": "<ISO timestamp>"}` to `.planning/workflow-memory.json`. Create file as `[]` if missing. Drop oldest entry if log length >= 50.

People who liked this also viewed

6 picks

Type