Cytation AIAggregated social feed with automatic fake content detection

ContentGuard AI

ContentGuard AI is a free browser tool that combines AI content detection, plagiarism, grammar checks, readability, and citation output.

ModelVerify.ai

ModelVerify.ai checks whether an LLM API actually serves the model it claims, comparing behavior against official fingerprints to catch swapped models, drift and latency issues.

Verol

Verol is a Chrome extension that adds an independent verification layer to ChatGPT, Claude, and Gemini. It parses answers, executes real-time web lookups, and validates sources via a dedicated backend pipeline, offering instant verdicts, confidence metrics, and clickable sources. No data tracking; history stays local. 5 free runs for testing, plans from $4.99/mo.

Fishy Email

Fishy Email is an AI phishing detection tool that works through email itself. Users do not need to install apps or extensions; simply forward suspicious emails to [email protected] and receive a full AI analysis in about 30 seconds. The analysis includes a color-coded verdict, a plain-English breakdown of each red flag, and recommended next steps. Every email address gets 5 free checks with no account required.

Axóncaptcha

Axóncaptcha is an advanced bot protection solution that uses real behavior detection to identify legitimate users. Unlike traditional methods, it analyzes how users type, not just what they type, enabling real-time blocking of bots, spam, and abuse. The product prioritizes user data privacy during verification and responds instantly to automated threats.

Humanizer: Open-Source Skill to Remove AI Writing Patterns

Humanizer is an open-source agent skill that removes common AI writing patterns from text to make output read as more human. It is distributed as a single Markdown SKILL.md file and works with multiple agent harnesses including Claude and Cursor. The skill catalogs 33 telltale patterns based on Wikipedia guidance on AI-generated writing, grouped into content, language, style, and communication categories. It also includes a final audit pass and supports voice calibration against a user-provided writing sample.

Avoid AI Writing: Audit and Rewrite AI Content for Natural Flow

Avoid AI Writing is an open-source JavaScript tool that audits and rewrites AI-generated content, stripping away tell-tale machine patterns. It integrates with AI agents like Claude Code and OpenClaw, helping users produce more natural, human-like articles. The project is licensed under MIT and has 2548 stars on GitHub (as per README).

UQLM: Uncertainty Quantification to Detect LLM Hallucinations

UQLM is an open-source Python package designed to detect hallucinations in large language models by quantifying their inherent uncertainty. It provides measurable confidence assessments for reliability-sensitive applications, helping developers and researchers mitigate risks associated with LLM-generated content. This tool offers a pragmatic approach to understanding when an LLM might be fabricating information, moving beyond simple confidence scores to a more robust uncertainty quantification framework.