factcheck

Extract factual claims, verify them via web search, and generate correction drafts.

11|1|Updated Apr 10, 2026
One-click install
npx skills add https://github.com/Arcanada-one/datarim --skill factcheck
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: factcheck
Source: https://github.com/Arcanada-one/datarim/tree/main/skills/factcheck
Command: npx skills add https://github.com/Arcanada-one/datarim --skill factcheck

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps prevent publishing inaccurate or misleading content by extracting verifiable claims, checking them against authoritative sources, and applying minimal corrections while preserving the author’s voice.

Core Features & Use Cases

  • Claim extraction: Identifies every verifiable factual claim with location and importance so nothing is skipped.
  • Structured verification: Uses a Chain of Verification approach (search → fetch authoritative pages → cross-reference) and assigns per-claim verdicts.
  • Minimal, consistency-safe edits: Corrects inaccuracies and resolves internal contradictions without rewriting the document’s style or structure.
  • Use case: Before publishing an article or post, you can run factcheck to catch outdated versions, wrong numbers, incorrect names, and missing context, then update only the affected statements.

Quick Start

Run factcheck on the article file by providing the path you want checked.

Frequently Asked Questions about factcheck

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

FAQPage Schema
How do I verify factual claims in an article before publishing?

To verify factual claims in an article, you can run an automated fact-checking process that extracts verifiable statements, searches authoritative web sources, and assigns per-claim verdicts. It then applies minimal corrections to fix inaccuracies without altering the original structure.

What is the best way to fact-check technical documentation for outdated information?

Fact-checking technical documentation involves extracting verifiable claims and cross-referencing them against fetched authoritative sources. This process identifies outdated versions or wrong numbers, generating a correction draft and a detailed report in a temporary working directory.

Can I fix content errors and resolve internal contradictions without rewriting the document?

Yes, you can fix content errors and resolve internal contradictions without rewriting the document by applying consistency-safe edits. This approach corrects only affected statements and validates claims while strictly preserving the author's original voice and structural formatting.

How does automated claim verification work for social posts and research writing?

Automated claim verification works by using a Chain of Verification approach: it first searches for sources, then fetches authoritative pages, and finally cross-references the data against the text. It assigns a verdict to each extracted claim to ensure content accuracy.

Does fact-checking content require providing source citations for every claim?

Fact-checking content requires fetching authoritative pages to confirm extracted claims, ensuring source citation validation. While it checks claims against these sources, it applies minimal text corrections and outputs a report rather than inserting inline citations into your document.

What should I expect as an output when checking an article for accuracy?

When checking an article for accuracy, the output includes a correction draft featuring minimal, consistency-safe edits and a verification report. These files are generated in a temporary working directory, detailing the extracted claims, their verdicts, and the applied corrections.