squeeze

Detect AI-generated text using Normalized Compression Distance against known corpora.

6|1|Updated Jun 23, 2026

Search Tags:#compression #text analysis #ai detection #text authenticity #ncd

Ready to run

npx skills add https://github.com/JGalego/Bag-of-Tricks --skill squeeze

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: squeeze
Source: https://github.com/JGalego/Bag-of-Tricks/tree/main/squeeze
Command: npx skills add https://github.com/JGalego/Bag-of-Tricks --skill squeeze

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires lzma, zlib, bz2, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill allows users to quickly determine if a piece of text is likely generated by an AI or a human, based on its compression properties.

Core Features & Use Cases

AI Text Detection: Assess whether text is AI-generated using Normalized Compression Distance (NCD).
Fast Analysis: Process and analyze text for AI-likeness with minimal delay.
Use Case: If you have a text that you suspect might be AI-generated, run it through the squeeze Skill to get an immediate assessment.

Quick Start

Analyze the text in the provided document 'analysis.txt' to determine if it's likely AI-generated.

Frequently Asked Questions about squeeze

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

FAQPage Schema

How do I detect if text is AI-generated using compression analysis?▼

AI-generated text detection via compression analysis evaluates text authenticity by comparing its Normalized Compression Distance against known AI and human corpora. This approach provides an immediate assessment of whether content is likely machine-generated based on its compression properties.

What is Normalized Compression Distance for text authenticity?▼

Normalized Compression Distance (NCD) is a metric used to assess text authenticity by measuring the similarity between a target text and reference corpora through compression ratios. It identifies AI-generated content by analyzing how efficiently the text compresses compared to known human and AI samples.

Do I need Python lzma or zlib modules to run AI text detection?▼

Yes, you need Python's lzma, zlib, or bz2 modules to perform AI text detection with this Skill. These compression libraries are required dependencies for calculating the Normalized Compression Distance and evaluating the likelihood of text being AI-generated.

Can I use compression-based AI detection for academic and administrative workflows?▼

Yes, compression-based AI detection is designed for rapid evaluation of text authenticity in administrative and academic workflows. It provides a fast analysis method to process text and assess whether it is likely AI-generated with minimal delay.

How do I analyze a document to check if it was written by AI?▼

To analyze a document for AI authorship, provide the text in a file like 'analysis.txt' and run it through the squeeze Skill. The tool calculates the text's compression distance against reference corpora to determine the likelihood of AI generation.

Best way to check text authenticity without dedicated machine learning models?▼

Using Normalized Compression Distance offers a straightforward way to check text authenticity without training dedicated machine learning models. By leveraging standard Python compression modules like lzma, zlib, or bz2, it compares text against known corpora to rapidly estimate AI-likeness.