blog-factcheck

Extract statistical claims from markdown posts and verify cited URLs.

Updated Apr 12, 2026
One-click install
npx skills add https://github.com/captaindevv/blogs --skill blog-factcheck-captaindevv
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: blog-factcheck
Source: https://github.com/captaindevv/blogs/tree/main/skills/blog-factcheck
Command: npx skills add https://github.com/captaindevv/blogs --skill blog-factcheck-captaindevv

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Detects and verifies statistical claims and source attributions in blog posts so authors and editors can trust the numbers and citations they publish.

Core Features & Use Cases

  • Claim extraction: Scans markdown posts for numbers, percentages, dollar amounts, multipliers, and named attributions and produces a structured claims list.
  • Source verification: Fetches cited URLs, searches for exact or paraphrased matches, and assigns confidence scores (VERIFIED, PARAPHRASE, WEAK, NOT FOUND, UNVERIFIED).
  • Reporting & remediation: Generates a verification report with counts, per-claim status, notes, and suggested search queries for uncited claims; integrates as an optional deep-check for blog-analyze.

Quick Start

Use the blog-factcheck skill to fact-check path/to/post.md and produce a verification report summarizing verified, paraphrase, weak, not-found, and unverified claims.

Frequently Asked Questions about blog-factcheck

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

FAQPage Schema
How do I verify statistics and source claims in a blog post?

Blog fact-checking extracts numeric values and cited URLs from markdown content, fetches the referenced pages via WebFetch, searches for exact or contextual matches, and assigns confidence scores to flag uncited or unsupported statistical claims.

Can I check uncited numeric claims and get help finding sources?

Yes, this fact-checking process flags uncited claims containing numbers, percentages, or dollar amounts and generates suggested search queries to help authors locate and attach appropriate supporting sources for remediation.

What confidence levels are used when fact-checking editorial content?

Fact-checking assigns confidence scores of VERIFIED, PARAPHRASE, WEAK, NOT FOUND, and UNVERIFIED to each extracted claim based on whether exact or contextual matches are found on the fetched referenced pages.

Does blog fact-checking work with paywalled sources or dynamic content?

Blog fact-checking fetches cited URLs via WebFetch but explicitly notes paywalls or dynamic content limitations in the verification report when referenced pages cannot be fully accessed or parsed for matching statistical data.

How many cited URLs can be verified in a single blog post check?

A single blog fact-check run verifies up to ten cited URLs per execution, extracting claims from markdown and fetching each referenced page to search for exact or contextual matches and generate a comprehensive verification report.