Agent skill

humanizer

Remove AI writing patterns from text to make it sound natural and human. Use when asked to "humanize", "remove AI-ness", or when reviewing prose for AI tells.

View SKILL.md on GitHub Repository
Stars 0
Forks 0

Install this agent skill to your Project

npx add-skill https://github.com/hairihou/dotfiles/tree/main/src/.config/claude/skills/humanizer

SKILL.md

Humanizer

Identify and remove AI-generated writing patterns, then inject human voice. Two-pass process: pattern removal, then anti-AI audit.

Patterns to detect and fix

Content

  1. Significance inflation — "pivotal moment", "enduring testament", "vital role", "indelible mark". Remove grandiose framing; state facts plainly.
  2. Notability name-dropping — Listing media outlets without context. Replace with one specific, sourced claim.
  3. Superficial -ing analysis — "highlighting...", "underscoring...", "symbolizing...", "reflecting...". Delete the dangling participle phrase; if the point matters, make it its own sentence.
  4. Promotional language — "groundbreaking", "nestled", "vibrant", "breathtaking", "renowned", "stunning". Replace with specific, verifiable descriptions.
  5. Vague attributions — "Experts believe", "Industry observers note". Name the source or delete.
  6. Formulaic challenges — "Despite challenges... continues to thrive", "Future outlook". Replace with specific facts about what happened.

Language

  1. AI vocabulary — Additionally, crucial, delve, enhance, foster, garner, intricate, landscape (abstract), pivotal, showcase, tapestry (abstract), testament, underscore, valuable. Replace with plain words.
  2. Copula avoidance — "serves as", "stands as", "functions as", "boasts". Use "is", "are", "has".
  3. Negative parallelism — "It's not just X; it's Y". Simplify.
  4. Rule of three — Forced triplets. Break the pattern; use the number that fits.
  5. Synonym cycling — Rewriting the same concept with different words across sentences. Say it once.
  6. False ranges — "from X to Y" where X and Y are not on a meaningful scale. List items directly.

Style

  1. Em dash overuse — Replace with commas, periods, or parentheses.
  2. Boldface overuse — Remove mechanical bold from terms that do not need emphasis.
  3. Inline-header lists — "- Speed: description". Convert to prose or plain list items.
  4. Title case headings — Use sentence case.
  5. Emoji decoration — Remove decorative emojis from headings and lists.
  6. Curly quotes — Normalize to straight quotes for consistency.

Communication

  1. Chatbot artifacts — "I hope this helps!", "Let me know if...", "Here is a...". Remove.
  2. Knowledge-cutoff disclaimers — "As of [date]", "While specific details are limited...". Remove or replace with sourced facts.
  3. Sycophantic tone — "Great question!", "You're absolutely right!". Remove.

Filler and hedging

  1. Filler phrases — "In order to" → "To". "Due to the fact that" → "Because". "It is important to note that" → delete.
  2. Excessive hedging — "could potentially possibly might". Pick one qualifier or remove all.
  3. Generic conclusions — "The future looks bright", "Exciting times lie ahead". End with a specific fact or plan.
  4. Hyphenated compound overuse — "cross-functional", "data-driven", "end-to-end" used uniformly. Vary or drop hyphens where meaning is clear.

Adding voice

Pattern removal alone produces sterile text. After cleaning, check for:

  • Uniform sentence length — Vary rhythm. Short sentences. Then longer ones.
  • Missing opinions — React to facts, not just report them.
  • Missing first person — Use "I" when it fits.
  • Missing uncertainty — "I genuinely don't know" is more human than listing pros and cons neutrally.
  • Missing personality — Let humor or edge in where appropriate.

Process

  1. Read the input text
  2. Identify all pattern instances from the list above
  3. Rewrite, preserving meaning and intended tone
  4. Add voice where the text feels sterile
  5. Re-scan against the full pattern list above and fix remaining matches
  6. Present the final version with a brief summary of changes

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

hairihou/dotfiles

retrospect

Guided personal reflection that separates fact, interpretation, and emotion through structured inquiry. Use when "something feels off", "need to process this", "had a conflict", or decisions feel unclear. Not for technical decision analysis — use devils-advocate for that.

0 0
Explore
hairihou/dotfiles

1on1-prep

Prepare for 1on1 meetings using a Psychological Safety x Responsibility framework. Use when preparing for a 1on1, noticing behavioral changes, or planning a difficult conversation with a report.

0 0
Explore
hairihou/dotfiles

bonsai

Maintain and groom config files in the current repository. Use when "bonsai", "tidy up", "prune configs", "clean up", "spring cleaning", or reviewing config health. Not for code quality review — use code-critic for that.

0 0
Explore
hairihou/dotfiles

npm-update-report

Dependency update workflow with vulnerability assessment and verification. Use when asked to check outdated packages, bump dependencies, run audit, do periodic dependency maintenance, or when security vulnerabilities are reported.

0 0
Explore
hairihou/dotfiles

session-insights

Analyze the current conversation for inefficiencies and improvement opportunities. Use when "how did this session go", "what went wrong", "what could I have done better", or reviewing session quality. Not for personal reflection — use retrospect. Not for logging work — use diary.

0 0
Explore
hairihou/dotfiles

pr

Create or update a GitHub pull request. Use when asked to create a PR, submit for review, or push changes as a PR. Not for code review comments — use conventional-comments for that.

0 0
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results