llm-detection

Best static AI text humanizer. Two research-grounded LLM-agnostic skills that make AI writing sound human and relatable. Nine levers, 50+ peer-reviewed sources, 2024-2026 detection literature.

[AAAI 2025 oral] Official repository of Imitate Before Detect: Aligning Machine Stylistic Preference for Machine-Revised Text Detection

Detect and eliminate AI writing patterns in your content. This Claude Code plugin performs multi-tier analysis of character patterns, language cues, structural issues, and voice authenticity. Auto-fix em dashes, smart quotes, and emojis. Keep documentation and prose sounding genuinely human.

This project aims to address this gap by conducting a systematic, controlled study of human versus LLM-generated text detectability using paired question–answer datasets. Rather than proposing a novel detection architecture, the focus is on analyzing detection robustness, failure modes, and the impact of adversarial humanization strategies.

한글 AI 글 윤문·탐지 Claude Code 플러그인 — AI가 쓴 한국어를 사람처럼 윤문하고 AI 작성 여부를 진단

AI detection on your hardware

Professional text refinement, AI detection, and style conversion services. 专业文本润色、AI检测和风格转换服务

6-class text authorship detection pipeline for human and LLM-generated text using TF-IDF, stylometric features, and stacked scikit-learn/LightGBM models for the MALTO Hackathon 2026 (F1: 0.9393).

Catch AI code mistakes before they ship — 50+ checks for hallucinated APIs, stub functions, hardcoded secrets, and SQL injection. SARIF output for GitHub Code Scanning.

Browser-based LLM stylometric fingerprinting

The official repository for our ACL 2025 paper, "Who Writes What: Unveiling the Impact of Author Roles on AI-generated Text Detection"

Proof of concept tool to bypass document replay technology (such as GPTZero).

Python tool for simple comparison check on generated code vs suspected generated code.

🎲 Detect whether a GitHub repo's code was likely written by an LLM. Zero dependencies. Scores repos 0-100 using commit velocity, session analysis, burst detection, message patterns, and project-scale plausibility.

Detects AI-generated essays using an ensemble of LightGBM, CatBoost, Naive Bayes, SGD, and Random Forest. Custom BPE tokenizer built with Hugging Face + TF-IDF vectorization with 3-5 word n-grams. Weighted soft-voting classifier.

Quantifying the linguistic fingerprint of large language models

Projects concerning LLMs, prompting, NLP, webscraping, data aquisition and dataset analysis.

A unified tool for testing and using LLM detectors

Detecting AI coding assistance in famous OSS contributors — style drift analysis across 10 developers