ai-hallucination-detector

Detect and quantify hallucinations in AI-generated text with source verification.

1|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/Wangtengyu/ai-detector-skills --skill ai-hallucination-detector
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-hallucination-detector
Source: https://github.com/Wangtengyu/ai-detector-skills/tree/main/ai-hallucination-detector
Command: npx skills add https://github.com/Wangtengyu/ai-detector-skills --skill ai-hallucination-detector

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

该技能用于检测AI生成内容中的幻觉、编造、数据失实和伪科学营销等问题,提供自动化检测、词库自动学习扩展、权威总结生成、出处核验和报告摘要等能力。适用于核对AI生成的报告、材料、数据内容,识别不良商家SEO/GEO中的伪科学术语滥用。

Core Features & Use Cases

  • 自动化检测与核验:对长文本进行数据、引用、范围、科学词汇等多维核验,以及短文本的专业词汇/复杂逻辑检测。
  • 自动化检测脚本与词库学习:内置脚本实现自动分析,词库可学习扩展以提升识别覆盖。
  • 权威总结与出处核验:生成权威性摘要,提供出处核验与可溯源的核验清单。
  • 报告输出:生成结构化的核验报告及简要的风险评估。

Quick Start

将需要检测的文本输入到检测器,即可生成结构化的风险报告和权威总结。

Frequently Asked Questions about ai-hallucination-detector

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

FAQPage Schema
How do I detect AI hallucinations and fabricated data in long-form reports?

Source verification checks citations and origins in AI-generated content to prevent the spread of fabrication. It provides a traceable verification checklist and authoritative summaries to ensure data-driven documents are accurate and credible.

Can I identify pseudo-science terminology in AI-generated marketing copy?

Yes, you can identify pseudo-science terminology in AI-generated marketing copy and SEO content. The detector uses an automatically expanding vocabulary to recognize and flag the misuse of scientific terms, ensuring your marketing materials remain scientifically accurate.

What is the best way to verify citations and scientific vocabulary in AI-generated content?

The best way to verify citations and scientific vocabulary is using an automated detection script that cross-references text against a learning vocabulary database. This process outputs a structured verification report detailing data accuracy, citation validity, and risk assessment.

Does automated hallucination detection work for both long-form reports and short texts?

Automated hallucination detection works for both long-form reports and short texts. For long documents, it performs multi-dimensional verification of data and citations, while for short texts, it focuses on detecting complex logic issues and specialized professional vocabulary.