ai-check

Analyze text for AI-generated content using signal scoring and mixed-authorship estimation.

Updated Jun 30, 2026
One-click install
npx skills add https://github.com/SunS1eep1ng/codex-dotfiles --skill ai-check-suns1eep1ng
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-check
Source: https://github.com/SunS1eep1ng/codex-dotfiles/tree/main/skills/ai-check
Command: npx skills add https://github.com/SunS1eep1ng/codex-dotfiles --skill ai-check-suns1eep1ng

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides a detailed forensic analysis of text to determine if it was generated by AI, helping users identify AI-generated content and assess its authenticity.

Core Features & Use Cases

  • AI Detection: Analyzes text for patterns indicative of AI generation.
  • Signal Scoring: Scores various signals such as perplexity, burstiness, hedge density, and more.
  • Mixed-Authorship Overlay: Estimates the proportion of AI-written content in a mixed-authorship scenario.
  • Use Case: When reviewing a piece of text, use this Skill to determine if it was written by AI and to what extent.

Quick Start

Run the ai-check skill on the text provided in the 'text-input.txt' file.

Frequently Asked Questions about ai-check

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

FAQPage Schema
How do I detect AI-generated content in a text document?

To detect AI-generated content, you can perform a forensic text analysis that scores signals like perplexity, burstiness, and hedge density to identify patterns indicative of AI generation. This approach estimates the authenticity and proportion of AI-written text.

What signals indicate text was written by AI?

AI-generated text often exhibits specific signals such as low perplexity, uniform burstiness, and high hedge density. Analyzing these patterns through forensic text analysis helps verify content authenticity and distinguish AI-written portions from human text.

Can I estimate the proportion of AI-written content in a mixed-authorship document?

Yes, you can estimate the proportion of AI-written content in mixed-authorship scenarios by applying a mixed-authorship overlay during forensic text analysis. This process evaluates signal scores across the text to determine the extent of AI generation.

How do I verify the authenticity of a text for content review?

To verify text authenticity for content review, conduct a forensic analysis that examines perplexity, burstiness, and hedge density. This detects AI-generated patterns and provides a detailed assessment of whether the content was AI-generated.

Do I need to understand AI detection literature to analyze text for AI generation?

Yes, analyzing text for AI-generated content requires an understanding of AI detection literature and the ability to interpret various signals. This knowledge is necessary to accurately assess signal scores like perplexity and burstiness.

What are the limitations of forensic text analysis for AI detection?

Forensic text analysis for AI detection relies on interpreting signals like perplexity and burstiness, which can vary. Limitations include the need to understand AI detection literature to accurately interpret these signals and avoid misjudging mixed-authorship content.