anti-ai-tone

Detect AI-tone patterns in Chinese writing and output severity, location, and rewrite suggestions.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/AllenZhou/MIRISE --skill anti-ai-tone
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: anti-ai-tone
Source: https://github.com/AllenZhou/MIRISE/tree/main/agents/uae-docs-pipeline/skills/anti-ai-tone
Command: npx skills add https://github.com/AllenZhou/MIRISE --skill anti-ai-tone

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

MIRISE content often ships AI-style phrasing that reveals machine generation. This skill identifies common AI-tone patterns in Chinese writing and surfaces actionable guidance to produce more natural, authentic text.

Core Features & Use Cases

  • Preload constraints: loads banned words and structural patterns to constrain generation before writing.
  • Scanning and scoring: analyzes drafts to produce structured findings (severity, location, rewrite suggestions) for QA.
  • Rewrite mode: can generate LLM-friendly prompts for self-rewrite to fix AI-tone issues in drafts.

Quick Start

Run the anti-ai-tone-scan.ts script on your draft to generate a scan report and, if needed, switch to rewrite mode to produce an AI-tone-free version.

Frequently Asked Questions about anti-ai-tone

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

FAQPage Schema
How do I detect AI-tone patterns in Chinese writing?

To detect AI-tone patterns in Chinese writing, run a scan script that analyzes drafts against banned words and structural patterns, generating structured findings with severity levels and precise locations for QA scoring.

What are common AI-tone signs in Chinese documents?

Common AI-tone signs in Chinese documents include specific banned words and repetitive structural patterns that reveal machine generation, which are cataloged in dedicated reference files for scanning and identification.

How do I rewrite Chinese content to remove AI phrasing?

You remove AI phrasing by switching to rewrite mode, which generates LLM-friendly prompts for self-rewrite based on the initial scan report, guiding authors to produce authentic, natural text revisions.

Can I constrain AI generation to prevent AI-tone before writing?

Yes, you can prevent AI-tone before writing by preloading banned words and structural pattern constraints into the generation process, ensuring the model avoids machine-style phrasing during initial content creation.

Does the AI-tone scan output structured QA scoring data?

The AI-tone scan outputs structured QA scoring data that includes severity levels, issue locations, and specific rewrite suggestions, providing a comprehensive rubric for evaluating content quality and authenticity.