ai-tells

Detect AI-generated text in drafts using banned phrase and structural pattern flags.

Updated Apr 18, 2026
One-click install
npx skills add https://github.com/00PZ/oopz-inc --skill ai-tells-00pz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-tells
Source: https://github.com/00PZ/oopz-inc/tree/main/companies/shoshin/skills/ai-tells
Command: npx skills add https://github.com/00PZ/oopz-inc --skill ai-tells-00pz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Detects AI-generated or AI-contaminated text in drafts to protect brand voice and quality.

Core Features & Use Cases

  • Pattern-based detection of banned phrases and structural tells.
  • Editor-friendly flags with category and pattern name, plus exact sentence location.
  • Integrated into standard editorial workflows to trigger a two-fire rejection.

Quick Start

Load ai-tells into your editor workflow and start scanning drafts for the defined tells.

Frequently Asked Questions about ai-tells

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

FAQPage Schema
How do I detect AI-generated text in editorial drafts?

To detect AI-generated text in editorial drafts, load the ai-tells Skill into your editor workflow to scan for banned phrases and structural tells, enforcing brand voice integrity during the review process.

How does pattern-based content detection work for brand voice enforcement?

Pattern-based content detection works by scanning drafts for structural tells and banned phrases, applying a two-fire rejection threshold that automatically flags drafts for rejection when two or more AI-generated patterns are identified.

Can I get rewritten text suggestions when AI patterns are flagged?

No, you cannot get rewritten text suggestions when AI patterns are flagged. The detection process identifies the pattern name and exact sentence location but does not provide rewritten suggestions, leaving corrections to the editor's discretion.

When do I need a two-fire rejection threshold for AI text detection?

You need a two-fire rejection threshold for AI text detection when enforcing strict brand integrity in editorial workflows, ensuring drafts are automatically rejected only after two or more AI-generated pattern matches are identified.

Does ai-tells require any dependencies to scan drafts for AI tells?

No, ai-tells does not require any dependencies to scan drafts for AI tells. It operates independently within your standard editorial workflow to identify AI-contaminated text and flag banned phrases without external components.