Does it work with code comments?

Yes, it specifically identifies and removes redundant function descriptions and self-evident comments that don't add value beyond what the code itself expresses.

How does the confidence score work?

It rates the cleanup on a scale from 0.0 to 1.0, where high scores (0.8-1.0) indicate that text is significantly cleaner with all technical meaning preserved and no slop remaining.

Will this skill delete important technical information?

No, the skill uses strict context-preservation rules and a confidence assessment to ensure critical specifications, formulas, error conditions, and architectural rationales are protected.

Can I preview changes before they are applied?

Yes, the skill supports a preview mode that shows exactly what text will be removed or modified before any changes are finalized.

AI slop refers to the verbose, repetitive, and often unnecessary conversational filler (like 'Certainly! I'd be happy to help!') that LLMs frequently include in their outputs.

AI Text Cleanup & Slop Removal

Name: AI Text Cleanup & Slop Removal
Author: v1truv1us

byv1truv1us

•

Productivity & Workflow

Removes AI-generated verbosity, conversational fillers, and redundant technical content to improve information density.

The Text Cleanup skill provides a systematic framework for stripping away AI-generated 'slop'—the verbose, repetitive, and overly polite filler often found in machine-generated responses. By targeting specific patterns like conversational preambles, redundant code comments, and obvious documentation openers, this skill helps developers and technical writers maintain high-signal communication. It includes built-in confidence scoring and context-preservation rules to ensure that while the fluff is removed, critical technical nuances, architectural decisions, and security warnings remain perfectly intact.

Key Features

01Advanced pattern matching for hedging language and verbose transitions

02Confidence scoring system to ensure technical accuracy and meaning preservation

032 GitHub stars

04Automated removal of AI conversational preambles and excessive politeness

05Context-aware pruning of redundant code comments and self-evident documentation

06Customizable cleanup modes ranging from conservative to aggressive

Use Cases

01Refining AI-generated documentation to increase readability and professional tone

02Cleaning up auto-generated code comments that merely restate function signatures

03Standardizing team communication by removing filler from PR descriptions and commit messages

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add v1truv1us/ai-eng-system text-cleanup

For use in Claude.ai and ChatGPT

Download Skill