humanizer
Humanizer
Rewrite or edit text to eliminate telltale signs of AI-generated writing. This skill is based on documented, research-backed patterns that distinguish LLM output from human writing.
The core problem: LLMs regress to the mean. They replace specific, unusual, nuanced facts with generic, positive-sounding descriptions. The subject becomes simultaneously less specific and more exaggerated — like a photograph fading into a blurry sketch while someone shouts louder and louder that it shows a uniquely important person.
Your job is to reverse that: make text specific, grounded, and plainspoken.
Vocabulary Replacements
LLMs dramatically overuse certain words. A single instance might be coincidence; a cluster is a clear signal. Replace these with plainer alternatives or restructure the sentence entirely.
More from milistu/agent-skills
apple-human-interface-guidelines
Apple Human Interface Guidelines (HIG) reference for designing iOS, iPadOS, macOS, tvOS, visionOS, watchOS apps. Covers UI components, layout, accessibility, typography, navigation, inputs, and platform technologies. Use when designing Apple platform UIs, reviewing SwiftUI/UIKit patterns, or applying HIG design principles to any app.
51challenge-me
Direct technical advisor mode for counting/estimation and architecture/design
27pr-message
Write a concise, high-signal GitHub pull request message that explains intent, impact, and risk
26conventional-comments
Format review comments following the Conventional Comments standard. Use when writing code review comments, PR feedback, RFC reviews, peer reviews, or when the user asks for help with review comments.
25conventional-commits
Generate commit messages following the Conventional Commits v1.0.0 specification with Angular convention types. Use when creating git commits, writing commit messages, or when the user asks for help with commit messages.
24pr-review
Analyze and explain a pull request to help review it effectively
22