BeClaude

ai-text-humaniser

New
Community RegistryGeneralby Adnan Mueller

Removes AI-sounding patterns from generated text

First seen 6/1/2026

Summary

This skill removes AI-sounding patterns from generated text, making it more natural and human-like.

  • It's useful for developers who need to polish AI-written content for blogs, documentation, or communications.

Overview

A collection of specialized AI skills and workflows for Claude Code. Features a comprehensive development pipeline for solo founders plus writing and documentation tools.

Installation

Add this marketplace to Claude Code:

code
/plugin marketplace add adnanmueller/am-dev-plugins

Then install any plugin:

code
/plugin install development-pipeline@am-dev-plugins
/plugin install ai-text-humaniser@am-dev-plugins
/plugin install website-copy-standards@am-dev-plugins
/plugin install feature-documenter@am-dev-plugins

Available Plugins

PluginCategoryDescription
development-pipelineworkflowFull development workflow with 11 specialized agents
ai-text-humaniserwritingRemoves AI-sounding patterns from generated text
website-copy-standardswritingSEO/GEO optimized website copywriting
feature-documenterdocumentationGenerates feature documentation from code analysis

Development Pipeline

The flagship plugin — a comprehensive, stack-agnostic development workflow designed for solo founders.

Pipeline Overview

code
Phase 0  │ BOOTSTRAP         │ Project setup, stack selection
         │                   │
Phase 1  │ PRODUCT MANAGER   │ Requirements, user stories, priorities
         │                   │
Phase 2a │ UX/UI DESIGNER    │ Design brief (lite) or design system (full)
Phase 2b │ ARCHITECT         │ Technical architecture, API contracts, ADRs
         │                   │ [Parallel execution]
         │                   │
         │ ══ APPROVAL GATE #1 ══
         │                   │
Phase 3a │ BACKEND           │ API implementation, business logic
Phase 3b │ FRONTEND          │ UI components, state management
Phase 3c │ QA SPECS          │ Test specifications [Auto-triggered]
         │                   │ [Parallel execution]
         │                   │
Phase 4  │ QA VALIDATION     │ Execute tests, validate quality
         │                   │
Phase 5  │ DEVOPS            │ Local setup → Infrastructure → CI/CD
         │                   │
Phase 6  │ SECURITY AUDIT    │ Comprehensive security review
         │                   │
         │ ══ APPROVAL GATE #2 ══

Specialized Agents

AgentPhaseDescription
Pipeline CoordinatorMain entry point, orchestrates workflow
Bootstrap0Project initialization and setup
Product Manager1Requirements and user stories
UX/UI Designer (Lite)2aQuick design brief for MVPs
UX/UI Designer (Full)2aComplete design system
System Architect2bTechnical architecture and APIs
Backend Engineer3aServer-side implementation
Frontend Engineer3bClient-side implementation
QA Specs3cTest specifications
QA Validation4Test execution and validation
DevOps5Deployment and infrastructure
Security Audit6Security assessment

Key Features

  • Parallel Execution: Phases 2a/2b and 3a/3b/3c run in parallel
  • Human Approval Gates: Critical checkpoints before implementation and deployment
  • Universal Handoff Schema: YAML frontmatter for dependency tracking and versioning
  • Security Embedded: Security considerations tracked through every phase
  • Stack Agnostic: Works with any technology stack

Default Technology Stack

  • Frontend: Next.js 14 + Tailwind CSS + Zustand
  • Backend: Python 3.11 + FastAPI
  • Database: PostgreSQL + Prisma ORM
  • Deployment: Vercel (frontend) + Railway (backend)
  • CI/CD: GitHub Actions

Project Scope Types

ScopeDocumentationQualitySpeed
MVPLiteBalancedFast
ProductionFullComprehensiveThorough
PrototypeMinimalFunctionalFastest

Quick Start

  1. Install the plugin: /plugin install development-pipeline@am-dev-plugins
  2. Say: "I want to build [your idea]"
  3. Claude runs the Bootstrap agent to set up your project
  4. Progress through phases with next or phase [N]
  5. Use status to see pipeline progress

Commands

CommandDescription
statusShow current pipeline status
nextProceed to next phase
phase [N]Jump to specific phase
approve [doc]Approve a document
historyShow recent activity

AI Text Humaniser

Transforms AI-generated text into natural, human-sounding prose by removing common AI patterns:

  • Filler phrases ("It's worth noting that...", "Let's dive in...")
  • Corporate buzzwords (leverage, utilize, synergy)
  • Overwrought metaphors and weak intensifiers
  • Hedging patterns and mechanical transitions

Website Copy Standards

Professional website copywriting with:

  • SEO Optimization: Keyword integration, meta descriptions
  • GEO (Generative Engine Optimization): AI-friendly structured content
  • Accessibility: WCAG compliance, screen reader optimization
  • Structured Data: JSON-LD schema templates (FAQ, Article, Product, Organization)

Includes reference guides for B2B/B2C methodology, landing pages, brand storytelling, and cognitive copywriting.


Feature Documenter

Generates comprehensive feature documentation from code analysis:

  • Template-based documentation generation
  • Simple and comprehensive output formats
  • Code-to-documentation mapping

Contributing

  1. Fork the repository
  2. Create a feature branch (git checkout -b feature/new-plugin)
  3. Add your plugin to plugins/ with a .claude-plugin/plugin.json
  4. Update the marketplace.json to include your plugin
  5. Open a Pull Request

Plugin Structure

code
plugins/your-plugin/
├── .claude-plugin/
│   └── plugin.json
├── SKILL.md
└── [additional files]

License

MIT License — Use freely in your projects.


Built for solo founders who want structure without bureaucracy.

Install & Usage

1
Add a marketplace
/plugin marketplace add <org/repo>
2
Install the plugin

Add the configuration to /plugin install ai-text-humaniser@<marketplace>

3
Manage with /plugin
/plugin

Use Cases

Humanize AI-generated blog posts before publishing to avoid detection.
Rewrite technical documentation to sound less robotic and more engaging.
Polish AI-written email drafts to sound like a real person wrote them.
Adjust marketing copy to remove repetitive or unnatural phrasing.
Transform AI-generated social media posts into authentic-sounding content.
Edit AI-written code comments to use natural language instead of jargon.

Usage Examples

1

/ai-text-humaniser Rewrite this paragraph to sound like a human wrote it: 'We are excited to announce the launch of our new product, which leverages cutting-edge technology to provide an unparalleled user experience.'

2

/ai-text-humaniser Make this text less AI-sounding: 'The solution utilizes advanced algorithms to optimize performance and deliver superior results.'

3

Humanize this email draft: 'We would like to inform you that your account has been successfully upgraded. Please feel free to reach out if you have any questions.'

View source on GitHub
writingtexthumanize

Security Audits

LicenseUnknownSourceWarnRepositoryPass

Frequently Asked Questions

What is ai-text-humaniser?

This skill removes AI-sounding patterns from generated text, making it more natural and human-like. It's useful for developers who need to polish AI-written content for blogs, documentation, or communications.

How to install ai-text-humaniser?

To install ai-text-humaniser: add a marketplace (/plugin marketplace add <org/repo>), then add the config to /plugin install ai-text-humaniser@<marketplace>. Finally, /plugin in Claude Code.

What is ai-text-humaniser best for?

ai-text-humaniser is a plugin categorized under General. It is designed for: writing, text, humanize. Created by Adnan Mueller.

What can I use ai-text-humaniser for?

ai-text-humaniser is useful for: Humanize AI-generated blog posts before publishing to avoid detection.; Rewrite technical documentation to sound less robotic and more engaging.; Polish AI-written email drafts to sound like a real person wrote them.; Adjust marketing copy to remove repetitive or unnatural phrasing.; Transform AI-generated social media posts into authentic-sounding content.; Edit AI-written code comments to use natural language instead of jargon..