OUBIGFA/De-AI-Prompt-Enhancer-Writer-Booster-SKILL

去AI味提示词-作家增强-SKILL

What it solves

This project provides a set of specialized prompts designed to remove the "AI flavor" from Chinese text. It addresses the common problem of AI-generated Chinese sounding robotic, overly formal, or following predictable templates, making the output more readable and natural by either mimicking a specific human author's style or applying strict anti-template constraints.

How it works

The system is packaged as "SKILL" files compatible with tools like Claude Code, Cursor, and Windsurf. It operates through two primary modes:

  • good-writing: A style reproduction engine that extracts "writing DNA" from a set of reference articles to mimic a specific author's voice, including their vocabulary, sentence rhythm, and emotional tone.
  • de-AI-writing: A "patch" mode that focuses on faithful rewriting. It applies a 24-point AI trace detection system and strict constraints (e.g., banning specific transition words and robotic sentence structures) to clean up AI-generated text without adding or altering core information.

Additionally, the project includes automation scripts (style_audit.js and style-lint.ps1) to audit the style of generated text.

Who it’s for

  • Writers and content creators who want AI-generated Chinese to sound more human.
  • Users of AI-powered IDEs (Cursor, Windsurf, Claude Code) who want to integrate writing style constraints into their workflow.
  • Translators and researchers who need natural-sounding Chinese summaries or translations that avoid AI clichés.

Highlights

  • Dual-Mode Approach: Offers both a creative style mimicry mode and a strict "de-AI" cleaning mode.
  • AI Trace Detection: Utilizes a 24-item detection system to identify and remove common AI linguistic fingerprints.
  • Faithful Rewriting: Ensures that the "de-AI" mode does not hallucinate or add external facts, maintaining information conservation.
  • Customizable: Users can replace the reference articles in the .writer/ directory to train the system on their own personal writing style.

Related

  • Project
    larashero3-dotcom/lieflat-less-ai-tone

    A research-backed AI-writing de-flavoring skill that detects and minimally rewrites AI-style patterns using measurable indicators from controlled experiments.

  • Project
    MrGeDiao/shuorenhua

    A Chinese‑first AI‑assisted editing skill that removes boiler‑plate and stiff phrasing while keeping facts, conditions and the author’s voice. Load the three markdown rule files into a chat model (Claude, Grok, ChatGPT, etc.) and ask it to rewrite or just flag a passage. Provides mini, lite and full installations, granular control over how much restructuring is allowed, and a benchmark of 123 cases showing zero hard failures.

  • Project
    KKKKhazix/human-writing

    A framework and toolset for AI agents to eliminate "AI-style" Chinese writing by enforcing human-like rhythms, factual substance, and structural checks.

  • Project
    Tomsawyerhu/Chinese-WebNovel-Skill

    A modular, retrieval‑augmented prompt framework for AI‑assisted Chinese web‑novel writing. It splits the workflow into a short router (`SKILL.md`) and many self‑contained sub‑skills (concept planning, opening, plot logic, dialogue, consistency review, etc.). Each sub‑skill ships tutorials, runtime rules, good/bad examples, and may use local novel corpora via a provided search script. The system routes a user’s brief idea to the appropriate module, pulls relevant example excerpts, composes a concise prompt, generates text, and finally runs a consistency review. Designed for Chinese novel authors, prompt engineers, and researchers interested in modular LLM writing pipelines.

  • Project
    Raymondhou0917/speak-human-tw

    A proofreading skill for AI agents that removes "AI flavor" from Traditional Chinese text and localizes it for Taiwan, ensuring content feels human-written and natural.