Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Logic

ASecurity

逻辑与统计核查(测量原件,只产 finding,不改稿):claim graph 的循环论证、断链、偷换条件、充分/必要条件错误、未声明假设;样本与 split、CV/grouping、泄漏防护、metric、uncertainty、多重比较、prior 的可复现性;以及声明-证据纪律的审查端——动词强度不得超过证据强度。findings 走 sci-paper.feedback.v1。可单独运行,也是 /sci-paper:paper-review 维度 C 的唯一来源。Use for: 逻辑检查, 论证是否成立, 统计方法学, 数据泄漏, 多重比较, 声明证据纪律, claim-evidence, over-claim 检查。

5 stars
0 votes
0 copies
0 views
Added 9/28/2026
ai-agents

Security Analysis

A100/100

Scanned 9/28/2026

Install to Claude Code

$npx -y skills add skymanbp/sci-paper --skill logic --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Logic?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Logic
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/skymanbp-logic/badge)](https://www.skillsdirectory.com/skills/skymanbp-logic)

More formats (shields.io, HTML) on the badges page.

Files
SKILL.md
---
name: logic
description: 逻辑与统计核查(测量原件,只产 finding,不改稿):claim graph 的循环论证、断链、偷换条件、充分/必要条件错误、未声明假设;样本与 split、CV/grouping、泄漏防护、metric、uncertainty、多重比较、prior 的可复现性;以及声明-证据纪律的审查端——动词强度不得超过证据强度。findings 走 sci-paper.feedback.v1。可单独运行,也是 /sci-paper:paper-review 维度 C 的唯一来源。Use for: 逻辑检查, 论证是否成立, 统计方法学, 数据泄漏, 多重比较, 声明证据纪律, claim-evidence, over-claim 检查。
disable-model-invocation: false
argument-hint: "<file_path> [--field <name>]"
---

# logic — 逻辑与统计核查(测量原件)

> **Normative authority:** `docs/SCIPAPER_STANDARD.md`。findings 一律用
> `sci-paper.feedback.v1` 的后果类别与 measurement state。
> `/sci-paper:paper` 定义"声明-证据纪律";本 skill 是它的**审查端**。

**这是测量原件:只产 finding,不改稿。** 措辞层面的弱化改写由组合 skill 路由到
`/sci-paper:de-ai` Pass 3,并受该 skill 的 Preserve List 约束——**不得为了降低
风险把证据绑定的 hedging 改强,也不得为避词改动数字、引用或 stance**。

## 0. 取证纪律

1. **每条声明当轮定位到证据。** 声明的支撑必须在本轮从正文数字、图、表或引用
   重新读到,记忆和旧报告只能帮助定位。
2. **grep 只定位。** 命中与未命中都要 Read 上下文确认语义。
3. **不能验证就标 unmeasured。** 无法访问数据、脚本或 companion 时显式说明缺口,
   不能因为查不到就当作支撑成立。
4. **不可得不等于零 finding。**

## 1. claim graph

从全文抽出经验声明与推理链,检查:

- **循环论证** —— 结论被用来支撑自己的前提。
- **断链** —— 中间步骤缺失,结论推不出来。
- **偷换条件** —— 前提在推理过程中被悄悄放宽或收紧。
- **充分/必要条件错误** —— 把"必要"当"充分"用,或反之。
- **未声明假设** —— 论证依赖但从未写出来的前提。

无效推理、错误外推或 unsupported conclusion 为 `integrity_blocker`。
合法但表达不清的逻辑连接通常是 `advisory`。

## 2. 统计方法学

逐项检查可复现性:

- 样本定义、split 方式、CV / grouping 策略;
- **泄漏防护** —— train/test 之间、特征构造与标签之间、同源样本跨 split;
- metric 定义与其适用前提;
- uncertainty 的定义、来源与样本量;
- 多重比较是否校正;
- prior(log-uniform / flat / informative)是否声明。

无效统计、数据泄漏为 `integrity_blocker`。

**边界(canonical home — 本条)**:本维度管**经验方法学**。估计量本身的
物理/信息论前提(iid / Gaussian / 有限矩的显式声明、CLT vs LDP regime、
Pinsker / Fano / Cramér-Rao / DPI 的适用条件)归 `/sci-paper:physics` P4,
不在此复述。

## 3. 声明-证据纪律

`/sci-paper:paper` "声明-证据纪律"的审查端:

- 每个经验声明有**正文内**数字 / 图 / 表 / 引用支撑;
- **动词强度不超过证据强度** —— `demonstrates` / `establishes` / `proves` 各自
  需要什么级别的证据,由标准定义;
- 模糊量级写成**有归属的数字或区间**,而不是 "substantially" / "much larger";
- `significantly` 出现时必须伴随检验或数字,否则是 claim 缺陷。

分类:

| 情形 | 类别 |
|---|---|
| 无支撑,或动词强度超过证据强度 | claim-evidence defect → `integrity_blocker` |
| 措辞层面可以更准的弱化建议 | `advisory` |

**这不是词表判定。** `significantly` 之所以被点名,是因为它宣称了一个统计结论;
`landscape` / `demonstrate` 这类词的语域问题走 `tools/deai_register.py` 的
corpus document frequency,不在此按词表标记。

## 4. 边界 —— 本 skill 不做什么

- **不做物理第一原则核查。** 量纲、守恒、宇称、渐近、代数再推导归
  `/sci-paper:physics`(见 §2 边界)。
- **不做叙事结构审查。** "读者能不能跟上"归 `/sci-paper:mainline`;本 skill 只问
  "论证是否成立"。
- **不做语域 / AI-ism 判定。** 归 `tools/ai_ism_lint.py` 与 `/sci-paper:de-ai`。
- **不做引用真实性核验。** 引用是否存在、是否被伪造归 `/sci-paper:paper-review`
  维度 F;本 skill 只问被引内容**是否支撑该句的推理**。
- **不改稿。** 见顶部。

## 5. 报告

```markdown
# Logic — Reasoning and Statistics Findings

**Target**: <file> | **Field**: <field or none>

## Coverage
| area | status | method / why not applicable |
|---|---|---|
| claim graph / 统计方法学 / 声明-证据 | measured / unmeasured / not_applicable | … |

## Claim–evidence ledger
| claim | file:line | verb strength | evidence | verdict |

## Findings
| id | kind | location | evidence | recommended action | disposition |
```

不打印 PASS/FAIL 行。终止态是每条声明都有明确判定或明确记录的缺口。

## 6. 反模式

- ❌ "统计看着标准,跳过" —— 泄漏几乎总是藏在被跳过的那一步。
- ❌ 把 `significantly` 当词表命中标记,而不是去查有没有伴随的检验。
- ❌ 为了消掉 over-claim 把证据绑定的 hedging 改强——那本身是新的 claim 缺陷。
- ❌ 把"论证不成立"和"读者跟不上"混为一谈(后者归 `mainline`)。
- ❌ 把 `unmeasured` 写成通过。
- ❌ 复述 `/sci-paper:physics` 或 `/sci-paper:paper-review` 维度 F 已拥有的检查项。

Attribution

skymanbpskymanbp
View sourceMore from skymanbp →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".

1074701 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

695601 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3351 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

691 votes

math-skill

A comprehensive mathematical reasoning skill for AI assistants — handles arithmetic to research-level problems with rigorous step-by-step reasoning, systematic verification, and transparent uncertainty handling

381 votes
View all in ai-agents →