ref-verify

Verify academic references in .docx files using Python and AI sub-agents.

Updated Feb 17, 2026
One-click install
npx skills add https://github.com/elad-refoua/claude-code-skills --skill ref-verify
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ref-verify
Source: https://github.com/elad-refoua/claude-code-skills/tree/main/ref-verify
Command: npx skills add https://github.com/elad-refoua/claude-code-skills --skill ref-verify

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python-docx, openpyxl, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the time-consuming and error-prone process of verifying the accuracy of references in academic papers, ensuring scholarly integrity and saving researchers significant manual effort.

Core Features & Use Cases

  • Double-Control Verification: Employs two independent rounds of AI sub-agents to cross-reference citations against online sources, increasing accuracy and reliability.
  • Comprehensive Output: Generates a detailed Excel table with parsed reference fields and verification status, a Zotero-compatible RIS file for citation managers, and a summary of identified issues.
  • Use Case: A researcher preparing a manuscript can use this Skill to automatically check every citation against its source, flagging any discrepancies in authors, years, titles, or publication details before submission, thereby preventing potential rejections or corrections.

Quick Start

Use the ref-verify skill to check the accuracy of all references in the document 'my_research_paper.docx'.

Frequently Asked Questions about ref-verify

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

FAQPage Schema
How do I verify academic references in a Word document automatically?

You can automatically verify academic references in a .docx file by using a hybrid Python and AI sub-agent approach that parses citations and performs double-control web searches to check factual accuracy against online sources.

What is double-control verification for checking citations?

Double-control verification employs two independent rounds of AI sub-agents to cross-reference extracted citations against online sources, increasing the accuracy and reliability of the reference validation process.

Can I export verified references to Zotero and Excel?

Yes, the verification process generates a detailed Excel table containing parsed reference fields and verification status, alongside a Zotero-compatible RIS file for direct import into citation managers.

Do I need Python to check the accuracy of references in my manuscript?

Yes, the extraction, parsing, and AI-driven validation of scholarly references rely on Python dependencies like python-docx to read .docx files and openpyxl to generate the final Excel output.

What is the best way to flag discrepancies in authors, years, or titles before journal submission?

The best way is to run an automated AI-driven validation check that cross-references every citation against its source, outputting a summary of identified issues to flag discrepancies in authors, years, titles, or publication details.

What are the limitations of using AI sub-agents for reference verification?

The process is limited to parsing references from .docx documents and depends on web search availability via Sonnet sub-agents, meaning it cannot verify offline sources or citations embedded in PDF or LaTeX files.