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

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

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.

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).

A measuring instrument for text, not an AI detector. Estimates statistical signals of machine-generated language and reports them with measured false-positive rates, broken down by text category. Abstains when the evidence doesn't hold. Never asserts authorship. Runs entirely locally; Spanish calibrated first.

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.

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

Browser-based LLM stylometric fingerprinting

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.

A unified tool for testing and using LLM detectors

🎲 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.

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

Stylometric analysis & LLM-authorship detection for prose and source code — Python, JS, TS, C, Go, Rust

Modern local web GUI for the MELD AI-generated-text detector. Calibrated per-genre thresholds, sentence-level heatmaps, batch scanning, REST API.

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