plumbeer-article-reference-finder

Finds independent external references for people, organizations, and claims mentioned in an article.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/marioplumbeer/skills --skill plumbeer-article-reference-finder-marioplumbeer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: plumbeer-article-reference-finder
Source: https://github.com/marioplumbeer/skills/tree/main/.agents/skills/plumbeer-article-reference-finder
Command: npx skills add https://github.com/marioplumbeer/skills --skill plumbeer-article-reference-finder-marioplumbeer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Verifying an article's claims and finding credible sources for every named person, organization, or statistic is slow manual research. This Skill reads an article from a URL, pasted text, or local file and produces a sourced reference table mapping each mention to an independently found link. ## Core Features & Use Cases - Scoped extraction: Confirms which categories to extract first — named people and organizations by default, with factual claims and existing citations as opt-in extras. - Wikipedia-first sourcing for people: Searches Wikipedia for every named person, verifies identity against article context to avoid wrong-person matches, and falls back to official bios or reputable profiles. - Primary-source preference for claims: Traces statistics and factual claims to the original study, dataset, or official report rather than secondary news coverage. - Use Case: Before publishing a blog post that mentions several researchers and cites a 2023 WHO statistic, run this Skill to get a table of Wikipedia links for each person and the primary WHO report for the statistic, ready to paste into a References section. ## Quick Start Find references for this article and list sources for every person and organization it mentions.

Frequently Asked Questions about plumbeer-article-reference-finder

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

FAQPage Schema
How do I find sources that back up an article?▼

Provide the article as a URL, pasted text, or local file and ask for references. The Skill extracts named people and organizations by default, searches for independent sources via web search, and returns a table mapping each mention to a verified link with a summary.

How to fact-check statistics cited in an article?▼

Opt into the factual claims category when prompted. The Skill searches for the original study, dataset, or official report behind each statistic rather than secondary news coverage, and notes discrepancies when sources report different numbers.

Can it read articles from PDF or DOCX files?▼

Yes. It reads Markdown and text files directly, and uses the bundled pdf or docx skills to extract text from those formats first. For URLs it uses WebFetch, and asks you to paste the text if a page is paywalled.

Does it just repeat the links already in the article?▼

No. The deliverable is independently found external references, not the article's own citations. Existing citations are tracked separately and only if you opt into that category, so the two are never mixed in the output.

What happens when no credible source exists for a claim?▼

The table row is kept with the reference marked as no credible source found rather than forcing a weak match. This applies after exhausting Wikipedia, official bios, and reputable news profiles for people, or primary sources for claims.

Why might a Wikipedia link point to the wrong person?▼

Shared names are the most common failure mode. The Skill verifies each Wikipedia match against context in the article such as occupation, organization, and time period before accepting it, and searches more narrowly when verification fails.