story-deslop

Scan web-novel text for AI-writing traces and rewrite it through six cleanup gates.

1|Updated May 16, 2026
One-click install
npx skills add https://github.com/LoveIiei/oh-story-opencode --skill story-deslop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: story-deslop
Source: https://github.com/LoveIiei/oh-story-opencode/tree/main/.opencode/skills/story-deslop
Command: npx skills add https://github.com/LoveIiei/oh-story-opencode --skill story-deslop

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It helps you clean up web-novel text that reads “too AI” by identifying common AI writing patterns and constraints, then rewriting them to sound more natural and human.

Core Features & Use Cases

  • AI-trace scanning & scoring: Detects banned words, sentence-pattern templates, dialogue tagging habits, rhythm issues, direct psychology telling, and “sublime/summary” endings, then produces an AI-Slop diagnosis report with an overall severity level.
  • Six-gate cleanup with guardrails: After you confirm, it performs bounded edits (total delete/change target not exceeding 15%) using a staged workflow: banned-word replacement, template de-rigging, Show-Don’t-Tell psychology externalization, rhythm disruption, dialogue naturalization, and ending de-sublimation.
  • Project-aware targeting: Uses the current project context (via .active-book) to determine which chapter/text to process, and respects .deslop-whitelist protected phrases by never modifying them.

Quick Start

Use story-deslop to process the chapter stored in your current project context and return a full diagnosis plus a rewritten “less AI” version.

Frequently Asked Questions about story-deslop

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

FAQPage Schema
How do I remove AI writing traces from web novel chapters?

Removing AI writing traces involves scanning web-novel text for banned words, uniform rhythm, and AI-style summary endings, then applying bounded edits across six ordered gates to restore a natural human voice and generate a diagnostic quality report.

What are common AI writing patterns to look for when novel editing?

Common AI writing patterns in novel editing include banned words, templated sentence patterns, rigid dialogue tags, rhythm uniformity, direct psychology telling, and summary endings that lack natural human variation.

How do I apply show don't tell to fix direct psychology telling in fiction?

To apply show don't tell for direct psychology telling, you externalize internal character states into observable actions, dialogue, and sensory details, replacing direct emotional explanations with externalized behavior during the editing workflow.

Can I protect specific phrases from being changed during style polishing?

Yes, you can protect specific phrases during style polishing by creating a designated whitelist protection file, ensuring the cleanup workflow respects these terms and never modifies them throughout the six-gate editing process.

How much text does an AI-slop cleanup workflow change when polishing fiction?

An AI-slop cleanup workflow applies bounded edits with a total delete and change target not exceeding fifteen percent of the text, ensuring the original narrative structure remains intact while removing detectable AI traces.

How does dialogue naturalization work for Chinese web fiction?

Dialogue naturalization for Chinese web fiction removes rigid AI-style dialogue tags and templated conversational patterns, rewriting character exchanges to sound spontaneous and natural within the bounded editing workflow.