skills.rest
shysssthanhtri
shysssthanhtriCOMMUNITY
Explore all skills by this author →
GitHub

doublecheck

Extract verifiable claims from AI-generated text and verify them via web searches.

Updated Mar 19, 2026
Search Tags:#fact-checking#verification#reporting#citations#web-search#ai-safety#hallucination-detection
One-click install
npx skills add https://github.com/shysssthanhtri/AI-tools --skill doublecheck-shysssthanhtri
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: doublecheck
Source: https://github.com/shysssthanhtri/AI-tools/tree/main/skills/doublecheck
Command: npx skills add https://github.com/shysssthanhtri/AI-tools --skill doublecheck-shysssthanhtri

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes assets (resource) components.

What problem does it solve?

Doublecheck reduces the risk of relying on AI-generated content by extracting verifiable claims, locating supporting or contradicting sources, and flagging hallucination patterns so humans can independently verify results before acting.

Core Features & Use Cases

  • Three-layer pipeline: self-audit claim extraction, targeted web-source verification, and adversarial review for hallucination patterns.
  • Inline and full reporting: fast inline verification summaries for active mode and full templated three-layer reports on demand or for high-risk content.
  • Use Case: Validate assistant responses that include statistics, legal citations, or technical claims before publishing or using them in decision-making.

Quick Start

Run doublecheck on the assistant's previous response and produce an inline verification summary with confidence ratings and source links.

Frequently Asked Questions about doublecheck

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

FAQPage Schema
How do I verify AI-generated text for factual accuracy before publishing?▼

Verifying AI-generated text involves extracting verifiable claims, running targeted web searches for supporting or contradicting primary sources, and applying adversarial hallucination checks to generate inline summaries or full reports with confidence ratings.

What is the best way to detect hallucinations in AI research summaries?▼

Hallucination detection in AI research summaries uses an adversarial review layer that cross-references extracted claims against located primary sources, flagging hallucination patterns so humans can independently verify the results before acting.

Can I fact-check legal citations and technical claims from conversational assistants?▼

Yes, fact-checking legal citations and technical claims from conversational assistants is possible using a three-layer verification pipeline that extracts claims, locates supporting or contradicting sources, and flags hallucination patterns for decision-making.

How does claim extraction work for fact-checking AI outputs?▼

Claim extraction for fact-checking AI outputs works by isolating verifiable statements from generated text, which enables targeted web searches to find supporting or contradicting primary sources for each specific extracted claim.

Does verification reporting provide inline summaries or full reports?▼

Verification reporting provides both fast inline verification summaries with confidence ratings for active mode and full templated three-layer reports on demand, specifically for validating high-risk content like legal or technical analyses.

Related Skills

Explore More Legal & Compliance Skills→
kishorkukrejakishorkukreja
56

nlp-supply-chain

Apply NLP techniques to supply chain text for extraction and classification.

Community
Advanced
mishahaninmishahanin
10

osint

Gather multi-source intelligence on companies, individuals, markets, and technologies.

Community
Advanced
angryconsultantangryconsultant

doublecheck

Verify claims in AI outputs and surface credible sources.

Community
Advanced
coff33ninjacoff33ninja

verify-and-cite

Verifies AI-generated claims with sources and expresses uncertainty in responses.

Community
Advanced