text-humanizer-Instruction-based

What This Skill Does

The text-humanizer-Instruction-based skill is a linguistic refinement engine designed to neutralize the distinctive patterns associated with Large Language Model (LLM) output. AI-generated text often suffers from repetitive structural habits, overused metaphors like 'delve' or 'landscape,' and a sterile, sycophantic tone that signals 'bot' to human readers. This skill performs a granular scan against a comprehensive master list of 45 specific AI linguistic tells. Instead of relying on external APIs, it utilizes the agent's internal reasoning capabilities to identify these patterns and apply semantic rewrites that restore a natural, human cadence to the prose. It is ideal for polishing marketing copy, academic abstracts, professional emails, and personal posts, effectively bridging the gap between machine-generated content and authentic human expression.

Installation

To install this skill, run the following command in your terminal: clawhub install openclaw/skills/skills/arbazex/text-humanizer-instruction-based Ensure you have the OpenClaw environment initialized before execution. The skill source is maintained in the openclaw/skills repository.

Use Cases

• Professional Communication: Revitalizing corporate emails or LinkedIn posts that read as overly corporate or AI-written.
• Content Marketing: Transforming blog posts that use excessive buzzwords into engaging, direct articles.
• Academic/Creative Writing: Smoothing out awkward punctuation patterns like over-reliance on em dashes or 'rule-of-three' list structures.
• Brand Consistency: Ensuring the brand voice sounds distinct and personal rather than algorithmically uniform.

Example Prompts

1. "I used ChatGPT to draft this article, but it sounds really robotic. Can you humanize this for me?"
2. "This cover letter sounds like it was written by an AI—it keeps saying 'delve' and 'landscape'. Please rewrite it to sound like I wrote it."
3. "Please fix the robot-speak in this social media post; it feels too formal and uses way too many em dashes."

Tips & Limitations

• Know the Context: Always review the output to ensure the 'humanized' version hasn't lost the specific intent of your original draft.
• Non-Technical Use: This skill is optimized for natural language, not technical documentation or code. Do not use it for codebase cleanup.
• Detection Accuracy: While highly effective at stripping common AI patterns, ensure your input does not contain highly specialized technical jargon that might be misidentified as a buzzword.
• Silent Execution: The skill will automatically apply fixes. If you require a list of detected patterns, please explicitly ask the agent for a 'detection report' after the rewrite.