humanize

What problem does it solve? AI-generated prose carries recognizable tells: inflated significance, promotional vocabulary, em dash overuse, chatbot artifacts, and metronomic rhythm. This Skill finds those patterns in drafts, files, or entire documentation folders and rewrites the text to sound natural without inventing facts. ## Core Features & Use Cases - 41-pattern detection catalog: Covers content, language, style, communication, filler, and rhetoric patterns derived from Wikipedia's Signs of AI writing guide, with false-positive guardrails so legitimate human prose is left alone. - Draft, audit, final rewrite loop: Produces a draft rewrite, audits it for remaining AI tells and fabricated facts, then delivers a final version with no em dashes and no invented claims. - Voice calibration: Analyzes a sample of your own writing and matches its sentence lengths, vocabulary, and quirks instead of producing generic clean output. - Repo auditing with sloplint: A zero-dependency Node CLI scores files 0-100, lists findings with line numbers, applies safe mechanical fixes, and gates CI with a fail-above threshold. - Use Case: Point it at a docs folder before a launch. It ranks files by how AI-flavored they are, you pick the worst offenders, and each rewrite lands as a reviewable git diff. ## Quick Start Ask the agent to humanize a pasted draft or a file, for example: humanize docs/launch-post.md and keep every fact from the original.