← Back to Skills Marketplace
harrylabsj

Argument Audit

by haidong · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
148
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install argument-audit
Description
论点审计 - An analytical tool that evaluates the logical structure, evidence quality, and persuasiveness of arguments. Use when user asks about 论点分析、论证评估、逻辑审查、观点...
README (SKILL.md)

Argument Audit (论点审计)

Overview

This skill provides systematic analysis and evaluation of arguments. It helps users understand the logical structure of arguments, assess the quality of evidence, identify logical fallacies, and evaluate overall persuasiveness. Useful for critical thinking, debate preparation, writing analysis, and decision-making.

When to Use This Skill

  • Analyzing essays, articles, or opinion pieces
  • Preparing for debates or discussions
  • Evaluating business proposals or recommendations
  • Reviewing academic arguments
  • Self-checking own arguments before presenting
  • Understanding propaganda or misleading content

What This Skill Analyzes

1. Logical Structure

  • Argument identification — Main claim vs. supporting points
  • Reasoning chains — How conclusions follow from premises
  • Argument type — Deductive, inductive, abductive reasoning
  • Structure flaws — Missing links, unsupported assumptions

2. Evidence Quality

  • Source credibility — Authority, expertise, potential bias
  • Evidence relevance — Direct support vs. tangential
  • Evidence sufficiency — Adequate support for claim
  • Data quality — Sample size, methodology, recency

3. Logical Fallacies

  • Ad hominem attacks
  • Straw man arguments
  • False dilemmas
  • Slippery slope
  • Circular reasoning
  • Appeal to authority/emotion
  • Hasty generalizations
  • Post hoc ergo propter hoc
  • Red herring
  • Tu quoque

4. Persuasiveness Assessment

  • Overall argument strength (1-10 scale)
  • Emotional vs. rational appeal balance
  • Counterargument consideration
  • Balance and objectivity

Audit Framework

Dimension Criteria Questions
Clarity Is the main argument clear? What exactly is being argued?
Validity Does the reasoning hold? Do conclusions follow from premises?
Soundness Are premises true? Is the evidence credible?
Completeness Are key points covered? What's missing?
Fairness Are alternatives considered? Are counterarguments addressed?

Workflow

  1. Input Analysis — Parse the argument into components
  2. Structure Mapping — Diagram the logical structure
  3. Evidence Review — Assess each piece of supporting evidence
  4. Fallacy Check — Scan for common logical fallacies
  5. Strength Rating — Evaluate overall persuasiveness
  6. Improvement Suggestions — Recommend ways to strengthen

Usage Examples

Analyzing Written Arguments

"帮我分析这篇文章的论点"
"这篇论文的论证有什么问题?"
"这个商业计划书的逻辑是否严密?"

Debate Preparation

"帮我找出对方论点中的漏洞"
"如何反驳这个观点?"
"我的论点有什么薄弱环节?"

Critical Evaluation

"这个说法有什么逻辑问题?"
"这个专家的观点可信吗?"
"这段内容有哪些隐藏的假设?"

Output Format

For each audit, provide:

## Argument Summary
[1-2 sentence description of the main argument]

## Structure Analysis
[Diagram or breakdown of reasoning structure]

## Evidence Assessment
| Evidence | Source | Credibility | Relevance | Rating |
|----------|--------|-------------|-----------|--------|
| ...      | ...    | ...         | ...       | ...    |

## Fallacy Detection
- [List any logical fallacies found with explanations]

## Strength Rating
- Clarity: X/10
- Validity: X/10
- Soundness: X/10
- Completeness: X/10
- Overall: X/10

## Recommendations
[Suggestions for strengthening the argument]

Limitations

  • Cannot access external sources to verify evidence claims
  • Assessment is based on provided text only
  • Context-dependent factors may not be fully captured
  • Subjective elements exist in persuasiveness ratings

Related Skills

  • summarize — For summarizing argument content
  • cognitive-reframe — For rephrasing arguments
  • decision-fatigue-reliever — For simplifying complex arguments
Usage Guidance
This skill appears internally consistent and works only on text you provide. Before installing, consider: (1) it cannot verify external facts or fetch sources — do not rely on it to validate claims against the web; (2) its fallacy detection uses brittle heuristics and may produce false positives or flag text that contains offensive words (those words are only in its detection lists); (3) the handler.py shown is local and contains no network/file I/O, but a full review of the complete file is recommended if you require higher assurance (the provided snippet ends with a truncated marker — obtain the full source if concerned). Avoid submitting sensitive secrets or private data to any third‑party skill even if the code appears local.
Capability Analysis
Type: OpenClaw Skill Name: argument-audit Version: 1.0.0 The 'argument-audit' skill is a legitimate analytical tool designed to evaluate logical structures and identify fallacies in text. The code in handler.py uses basic keyword matching and regex to perform its analysis without any network access, file system manipulation, or data exfiltration, and the SKILL.md instructions are strictly aligned with the tool's stated purpose.
Capability Assessment
Purpose & Capability
Name/description align with the included handler.py and SKILL.md. The skill performs local analysis of argument text (structure, evidence, fallacies) and does not request unrelated capabilities, binaries, or credentials.
Instruction Scope
SKILL.md instructs the agent to analyze user-provided text only; handler.py implements local text analysis and fallacy detection and does not call external services, read files, or access environment variables. Note: the detection heuristics include hard-coded keyword lists (some offensive terms used only to detect ad hominem) and simple pattern rules that can produce false positives/negatives.
Install Mechanism
No install spec provided (instruction-only skill). There are no download/extract/install steps or third-party package pulls in the metadata.
Credentials
The skill declares no required environment variables, no credentials, and no config paths — which is appropriate for a purely local text-analysis utility.
Persistence & Privilege
No elevated persistence requested (always:false). The skill does not request system-wide changes or modification of other skills' configs.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install argument-audit
  3. After installation, invoke the skill by name or use /argument-audit
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of argument-audit. - Provides systematic analysis of arguments, including logical structure, evidence quality, fallacy detection, and persuasiveness assessment. - Offers a step-by-step audit workflow and detailed output format for clarity and usability. - Supports use cases such as essay analysis, debate prep, business proposal review, and critical evaluation. - Includes structured frameworks and rating criteria to guide consistent argument evaluation. - Lists key limitations and related skills for user guidance.
Metadata
Slug argument-audit
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Argument Audit?

论点审计 - An analytical tool that evaluates the logical structure, evidence quality, and persuasiveness of arguments. Use when user asks about 论点分析、论证评估、逻辑审查、观点... It is an AI Agent Skill for Claude Code / OpenClaw, with 148 downloads so far.

How do I install Argument Audit?

Run "/install argument-audit" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Argument Audit free?

Yes, Argument Audit is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Argument Audit support?

Argument Audit is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Argument Audit?

It is built and maintained by haidong (@harrylabsj); the current version is v1.0.0.

💬 Comments