ai-hallucination-fact-check-protocol

Community

Verify AI citations with an adapted SIFT

AuthorGarethManning
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill helps students avoid being misled by AI-generated citations and statistics by providing a repeatable protocol for checking whether claims are real and correctly represented.

Core Features & Use Cases

  • AI-adapted SIFT: Replaces the usual “Investigate the source” step with AI-specific source reconstruction (does the cited source exist, and does it say what the AI claims?).
  • Hallucination taxonomy: Classifies common LLM error types such as fabricated citations, statistical invention, misattribution, and false consensus so verification steps match the claim type.
  • Classroom-ready activity: Includes a structured “Hallucination Hunt” plus a teacher think-aloud model to demonstrate both successful verification and discovery of fabrication.

Quick Start

Use this Skill for a fact-checking lesson when you have an AI response containing citations or numbers and want students to verify them using source reconstruction steps.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: ai-hallucination-fact-check-protocol
Download link: https://github.com/GarethManning/education-agent-skills/archive/main.zip#ai-hallucination-fact-check-protocol

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.