skills.rest
lynote-ailynote-aiOfficial·2 Agent Skills Included

ai-detector-skill

Explainable AI-generated text risk scoring with evidence signals

Analyzes essays, emails, articles, and reviews to estimate how likely the text was AI-generated. Returns a 0-100 risk score with confidence level, strongest evidence signals, and honest caveats instead of overconfident verdicts. Refuses to overstate weak evidence on short samples and recommends human review for high-stakes decisions. Runs fully offline through a local CLI and Python API with reproducible benchmark scripts.
433|15
↗
npx skills add lynote-ai/ai-detector-skill --all -g -y
Available:

Tells your AI agent when to run the detector, how to execute the CLI command, and how to phrase cautious, non-accusatory results with scores, signals, and caveats.

All Skills in This Repository (2)

Pure Emerald Level Indicators
📦 In Repo
lynote-ailynote-ai

ai-text-detector-github

Detect AI-like text signals and estimate risk in writing samples.

Official
Intermediate
📦 In Repo
lynote-ailynote-ai

ai-detector

Analyze text files for AI-generated content risk using the aidetect Python package.

Official
Intermediate

Frequently Asked Questions

FAQPage Schema
How to install ai-detector-skill?▼

Run `npx skills add lynote-ai/ai-detector-skill --all -g -y` in your terminal to install the skill globally for your agent.

How to check if text was written by AI?▼

Pass any essay, email, or article to the detector and it returns a 0-100 risk score with the strongest evidence signals and confidence level. It works best on samples longer than about 80 words.

Is the AI detection score proof of cheating?▼

No. The score is a cautious risk estimate, not proof, and the tool is designed to never accuse anyone of misconduct. High-stakes decisions should always include human review and comparison with known writing samples.

Does ai-detector-skill work offline?▼

Yes. It runs entirely locally with no network calls, using explainable weighted signals through a CLI or Python API.

Does it work with Claude Code and Codex?▼

Yes. It ships as a standard SKILL.md package with an AGENTS.md orchestrator, so repo-aware agents like Claude Code and Codex can use it directly.

Related Repositories in Content & Communication

View All in Content & Communication→
anthropicsanthropics
172.9k

skills

Official Anthropic skills for documents, design, and developer workflows

Official📦 18 Skills
bladerblader
39.4k

humanizer

Rewrite AI-sounding text so it reads like a person wrote it

Community📦 2 Skills
DevolutionsDevolutions
25.9k

UniGetUI

Automate software translation reviews, diffs, and localization quality checks

Official📦 7 Skills