What problem does it solve? AI-generated text carries recognizable patterns—significance inflation, promotional language, em dash overuse, formulaic structures—that make writing sound robotic and erode reader trust. This Skill identifies 29 documented AI writing patterns and rewrites text to sound natural and human. ## Core Features & Use Cases - Pattern Detection and Removal: Scans for 29 documented AI tells including vague attributions, rule-of-three overuse, copula avoidance, sycophantic tone, and filler phrases, then rewrites each problematic section. - Voice Calibration: Analyzes a user-provided writing sample (inline or by file path) and matches its sentence rhythm, word choice, and punctuation habits in the rewrite. - Personality Injection: Goes beyond removing bad patterns by adding opinions, varied rhythm, first-person perspective, and specific details so the text has a genuine human voice. - Use Case: You drafted a blog post with an AI assistant and it reads like generic slop. Paste the text in, optionally attach a sample of your own past writing, and receive a draft rewrite, a self-audit of remaining AI tells, and a final polished version. ## Quick Start Humanize this blog post draft so it sounds like a real person wrote it, and match the voice in my sample file at notes/my-style.md.
