Fact-Check Skill
by petar-nauka
Fact-Check Skill is a semi-automated fact-checking and disinformation-detection skill for Claude and Codex that makes an AI verify a claim before it agrees with it — the practical fix for chatbots that would rather flatter you than check. Instead of a quick yes, it runs an 11-step analysis pipeline: claim decomposition, multi-language search, lateral reading, origin tracing, red-flag detection (40+ markers across six categories) and an MFS severity score. It combines four established methods — the SIFT methodology, the CRAAP test, prebunking/inoculation science and claim decomposition — and outputs a visual HTML 'Fact-Check Card' with a dark/light, mobile-friendly layout. It runs in four modes: full standard fact-check, source comparison, prebunking briefings, and a quick yes/no verdict. It even works on Claude.ai without web tools via a fallback mode, and includes an educational component that teaches lateral reading and how to spot manipulation techniques. It's MIT-licensed and needs no MCP servers.
Installation
Key Features
- ✓11-step analysis pipeline: claim decomposition, multi-language search, lateral reading and origin tracing
- ✓Red-flag detection with 40+ markers across six categories, plus an MFS severity score
- ✓Combines SIFT, the CRAAP test, prebunking/inoculation science and claim decomposition
- ✓Four modes: full fact-check, source comparison, prebunking briefing, and quick yes/no verdict
- ✓Visual HTML Fact-Check Cards with dark/light themes and a mobile-friendly layout
- ✓Works on Claude.ai without web tools via a fallback mode; MIT-licensed, no MCP servers required
Use Cases
- →Verifying a claim before publishing instead of trusting an AI that's trained to agree with you
- →Comparing how different sources report the same event to spot spin or fabrication
- →Building a prebunking briefing to inoculate a team against a viral piece of misinformation
- →Adding a verification step to your workflow so a confident hallucination gets caught, not shipped