An AI text humanizer that strips robotic patterns from prose, code, and academic writing to restore a natural human voice.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install slopbuster
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install slopbuster using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is slopbuster?
Slopbuster is a professional-grade tool designed to identify and remove the telltale signs of AI-generated content, often referred to as slop. It goes beyond simple grammar editing by performing a voice transplant, replacing sterile and formulaic LLM patterns with the varied rhythm, specificity, and opinion characteristic of human writing. By integrating these Openclaw Skills into your workflow, you can ensure that your documentation, source code, and creative writing maintain a high level of authenticity and engagement.
The tool is built on a foundation of analyzing thousands of content samples and peer-reviewed research on LLM detection. It addresses everything from significance inflation in marketing copy to tautological comments in source code. Whether you need a quick polish or a deep voice calibration against a specific author's style, slopbuster provides a structured, multi-pass approach to humanizing digital text.
slopbuster Use Cases
- Cleaning up AI-generated code comments and verbose naming conventions in software projects.
- Humanizing academic abstracts and research papers to remove formulaic transitions and significance filler.
- Refining marketing blog posts and emails to eliminate common AI vocabulary like delve, tapestry, and landscape.
- Auditing commit messages to ensure they are concise, active, and technically meaningful.
- Calibrating AI-generated drafts to match a specific brand voice or personal writing style.
How slopbuster Works
- Diagnose: The skill reads the input and loads relevant rule files for text, code, or academic modes to identify matching AI patterns.
- Rewrite: It applies pattern removals and injects human voice markers, such as contractions, active voice, and honest uncertainty.
- Two-Pass Audit: After the initial rewrite, the agent performs a second pass to identify any lingering robotic tells that may have been introduced during the first phase.
- Score and Report: The final output is scored on a 0-10 human-ness scale, and a detailed changelog is generated to explain every modification.
slopbuster Setup
To use slopbuster within your environment, you can invoke it directly via the CLI with various flags to control depth and mode:
# Standard humanization for a markdown file
/slopbuster article.md
# Target source code specifically to clean up comments and naming
/slopbuster src/main.ts --mode code
# Run a deep analysis calibrated against a human writing sample
/slopbuster draft.md --depth deep --voice-sample author.md
slopbuster Data Schema & Taxonomy
Slopbuster utilizes a structured taxonomy of rules and scoring to evaluate content quality:
| Component | Description |
|---|---|
| Rule Sets | Specialized markdown files containing patterns for text (24 rules), code (80+ patterns), and academic writing (49 rules). |
| Human-ness Scale | A 0-10 score where 0-3 is obviously AI and 8-10 is indistinguishable from a skilled human writer. |
| Tiered Scoring | Weighted detection system: Tier 1 (AI-specific tells), Tier 2 (Corporate buzzwords), and Tier 3 (Weak signals). |
| Output Reports | Includes the original score, draft rewrites, a list of remaining tells, and a final humanized version. |
slopbuster Advanced Features
- Voice Calibration: Deep mode allows the agent to analyze a human-written sample and match the target output to that specific style.
- Academic Integrity: A specialized mode for research papers that preserves disciplinary conventions while removing structural AI padding.
- Code Quality Audit: Scans source files for 16 structural code tells, including defensive null-checks and tutorial-style comments.
- Preservation Mode: A flag to maintain formal language in business documents while still stripping away the most egregious AI cliches.
- Multi-Pass Refinement: A critical two-step auditing process that prevents the creation of sterile, voiceless text during the humanization phase.
SKILL.md
Loading
Related Openclaw Skills
Codegirl Skill Creator
A sophisticated tool designed to distill communication history and coding activity into a persistent, persona-driven AI programming partner.
Diagram Cog
Diagram Cog transforms natural language descriptions into professional, interactive diagrams and high-quality PDFs without manual dragging or layout adjustments.
Context Memoize Skill
A high-performance caching utility designed to store and reuse frequently accessed context fragments to minimize AI processing overhead.
GIF Cog
GIF Cog is an AI-powered animation engine that creates seamless, platform-optimized GIFs and cinemagraphs from scratch using a multi-model production pipeline.
Clawder
Clawder is a production-grade AI coding skill that implements JT directives to eliminate agent laziness through mandatory verification and autonomous bug fixing.
ZhangDi Avatar
A high-fidelity digital persona of entrepreneur ZhangDi designed for strategic business analysis and decision-making using structured frameworks.
Featured*
