skills.rest
arbazex
arbazexCOMMUNITY
Explore all skills by this author →
GitHub

writing-credibility-auditor

Audit text for logical fallacies, unsupported claims, weasel words, and misleading statistics.

Updated Mar 21, 2026
Search Tags:#critical thinking#text analysis#fact-check#logical fallacies#credibility audit#weasel words#misleading statistics
One-click install
npx skills add https://github.com/arbazex/ai-agent-skills --skill writing-credibility-auditor-arbazex
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-credibility-auditor
Source: https://github.com/arbazex/ai-agent-skills/tree/main/research%20and%20analysis/writing-credibility-auditor
Command: npx skills add https://github.com/arbazex/ai-agent-skills --skill writing-credibility-auditor-arbazex

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Writers, editors, and researchers often need to evaluate whether a piece of text is logically sound and well-supported before citing, publishing, or acting on it, but manually checking for fallacies, vague language, and statistical misuse is slow and error-prone. ## Core Features & Use Cases - Four-Dimension Audit: Detects 12 named logical fallacies, 8 types of unsupported claims, 3 categories of weasel words, and 6 types of misleading statistics, each with exact excerpts and severity ratings. - Structured Credibility Report: Produces a scorecard, per-finding explanations with suggested fixes, observed strengths, and an overall credibility rating. - Use Case: A journalist pastes a draft op-ed and asks for a credibility audit; the skill flags an Appeal to Authority, a missing sample size, and a relative-risk statistic without a baseline, then suggests concrete corrections. ## Quick Start Paste your text and ask the agent to audit it for credibility, logical fallacies, weasel words, and misleading statistics.

Frequently Asked Questions about writing-credibility-auditor

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check a text for logical fallacies?▼

Paste the text and ask the agent to audit it for logical fallacies. The skill scans for 12 named fallacies including Ad Hominem, Straw Man, False Dilemma, and Post Hoc, returning each flagged excerpt with an explanation and suggested fix.

How to detect weasel words and unsupported claims in writing?▼

Ask the agent to check the text for weasel words or unsupported claims. The skill flags anonymous authorities like "experts say", vague quantifiers like "many studies", and claims lacking sample sizes, citations, or verifiable sources.

Can this skill fact-check claims against live sources?▼

No, the skill audits reasoning structure and language only, not real-world factual accuracy. It flags claims as unsupported within the text but does not verify them against external data; live verification requires a search tool or subject-matter expert.

Does the credibility audit require API keys or internet access?▼

No, the skill runs entirely on the agent's reasoning using the detection frameworks defined in SKILL.md. It requires no API keys, external tools, scripts, or internet access.

Can I run only one part of the credibility audit?▼

Yes, specify a single dimension such as "just check this for weasel words" and the skill runs only that scan. It then produces a simplified report covering only the requested dimension instead of the full four-part scorecard.

What are the limitations of automated credibility auditing?▼

The audit works best on English text and evaluates reasoning structure rather than domain-specific factual accuracy. Very short texts may return few findings due to limited content, and technical claims may still require expert verification.

Related Skills

Explore More Education & Research Skills→
dapperdiversdapperdivers
1

web-fetch

Fetch web pages and extract readable plain-text content from HTML.

Community
Basic
Bhavya-DhootBhavya-Dhoot

aras-coherence-reviser

Revise LaTeX manuscripts by integrating reviewer feedback and improving coherence.

Community
Advanced
megaeth-labsmegaeth-labs
4

doc-correctness

Verify documentation claims against source code, configs, and deployments.

Official
Advanced
TheViziusGroupTheViziusGroup
1

prose-technical-mechanics

Analyzes and edits prose using diction, syntax, rhythm, and argumentation mechanics.

Official
Intermediate