When in the workflow should I use this skill?

It is designed to be used as the final quality gate after a text transformation (like an G6 humanization) but before the final export or submission of a document.

What happens if a citation is missing after humanization?

The skill triggers a 'Critical Failure' status, flagging the missing citation and recommending that the transformation be rejected to maintain academic integrity.

What is the primary purpose of the f5 skill?

The f5 skill compares a refined or 'humanized' version of a text against the original to ensure that stylistic changes did not compromise citations, numerical data, or the intended meaning.

Does it measure how much the writing quality improved?

Yes, it provides a comparative analysis of writing quality scores and the reduction of detectable AI patterns between the original and the humanized versions.

Can it detect if a p-value or statistical result was changed?

Yes, it performs a rigorous Statistical Accuracy Check that flags any changes to numerical values, including p-values, N-sizes, percentages, and test statistics.

Academic Humanization Verifier

Name: Academic Humanization Verifier
Author: HosungYou

byHosungYou

0•

Learning & Documentation

Validates that AI-to-human text transformations preserve academic integrity, citations, and statistical accuracy.

The F5 Humanization Verifier is a specialized quality assurance tool designed to audit text after it has undergone stylistic refinement or 'humanization.' It acts as a final gatekeeper in the academic writing pipeline, performing rigorous cross-checks between original and modified drafts to ensure that critical citations are preserved, numerical data—such as p-values and effect sizes—remain unchanged, and the original scholarly meaning is intact. This skill is essential for researchers and academics who need to improve prose flow while maintaining the absolute precision and objectivity required for peer-reviewed publication.

Key Features

01Meaning Preservation Analysis: Uses semantic comparison to confirm that core claims and findings remain unchanged.

02Citation Integrity Check: Ensures every reference is preserved in the correct format and logical context.

03Statistical Accuracy Validation: Cross-references p-values, sample sizes, and test statistics to prevent data corruption.

04Writing Quality Scoring: Quantifies improvements in readability and reduction in detectable AI writing patterns.

05Academic Tone Audit: Verifies that the scholarly register and technical precision are maintained throughout the text.

060 GitHub stars

Use Cases

01Auditing transformed abstracts to ensure high-density information remains accurate and objective.

02Finalizing research papers after applying stylistic improvements to ensure no data was altered.

03Verifying that automated editing tools haven't accidentally removed or misattributed citations.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add hosungyou/diverga f5

For use in Claude.ai and ChatGPT

Download Skill