home/dot_claude/exact_skills/humanizer/SKILL.md
Identifies and removes AI writing patterns from text to restore natural, human voice. Use when user mentions humanize, remove AI patterns, make it sound human, natural writing, AI writing cleanup, or deAI.
npx skillsauth add guicaulada/dotfiles humanizerInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Rewrites text to remove characteristic AI writing patterns and restore natural human voice. Based on Wikipedia's "Signs of AI writing" catalog (WikiProject AI Cleanup), covering 24 patterns across five categories.
| Category | Patterns | Key Signals | |--------------------|----------|----------------------------------------------------------------------------------| | Content | 1-6 | Significance inflation, promotional tone, vague attribution, formulaic structure | | Language & Grammar | 7-12 | Overused AI vocabulary, copula avoidance, synonym cycling, false ranges | | Style | 13-18 | Em dash overuse, mechanical boldface, emoji decoration, title case headings | | Communication | 19-21 | Chatbot artifacts, knowledge-cutoff disclaimers, sycophantic tone | | Filler & Hedging | 22-24 | Filler phrases, excessive hedging, generic positive conclusions |
Full pattern reference with before/after examples: patterns.md.
Trigger: "humanize", "remove AI patterns", "make it sound human", "natural writing", "deAI"
Read and follow humanize.md.
testing
Fetches GitHub issues, pull requests, reviews, notifications, and project board data for the user and optionally teammates, then analyzes their state and produces a prioritized work plan. Use when prioritizing work, triaging GitHub activity, planning what to work on next, assessing current workload, checking sprint status, or reviewing project board items. Supports teammate usernames and focus areas via arguments. Do NOT use for creating, modifying, or closing issues, PRs, or other GitHub resources.
testing
Transforms text into exhaustive structured bullet-point summaries for studying and test preparation. Use when user mentions summarize, create a summary, bullet point this, give me the key points, study notes, help me study, prepare for exam, create study guide, or make flashcard content.
development
Reviews pull requests with detailed code analysis and feedback. Use when user mentions review pr, pr review, review pull request, check pr, analyze pr, code review, review prs, batch review. Not for dependency bump PRs (use /review-deps instead).
testing
Reviews dependency bump PRs with focus on compatibility and security issues. Use when user mentions review deps, review dependencies, dependency review, check deps, dependency prs, dep bumps, review deps prs, batch dep review.